Podcasts about unos

CardioNerds (Dr. Jenna Skowronski [Heart Failure Council Chair], Dr. Shazli Khan, and Dr. Josh Longinow) are joined by renowned leaders in the field of AHFTC (Advanced Heart Failure and Transplant Cardiology) and mechanical circulatory support, Dr. Jeff Teuteberg and Dr. Mani Daneshmand to continue the discussion of advanced heart failure therapies by taking a deep dive into the world of durable LVADs (Left Ventricular Assist Devices). In this episode, we will review the history of ventricular assist devices, the basics of LVAD function, selection criteria for LVAD therapy, and surgical nuances of LVAD implantation. Audio Editing by CardioNerds intern, Joshua Khorsandi. Enjoy this Circulation 2022 Paths to Discovery article to learn about the CardioNerds story, mission, and values. CardioNerds Heart Success Series PageCardioNerds Episode PageCardioNerds AcademyCardionerds Healy Honor Roll CardioNerds Journal ClubSubscribe to The Heartbeat Newsletter!Check out CardioNerds SWAG!Become a CardioNerds Patron! Pearls There have been significant advances in the field of MCS/LVAD therapy since the first implanted LVAD in the 1960s, to the first FDA approved device in the early 2000's, to now the HM3 LVAD, with the most important change being a centrifugal flow/magnetically levitated design that led to minimized hemocompatibility-related adverse events (HRAE's) (MOMENTUM 3 trial comparing HM2 and HM3).  The REMATCH trial in 2001 was a pivotal trial for LVAD therapy, demonstrating that in a population of patients with advanced HF (70% IV inotrope dependent), LVAD therapy significantly improved survival at both 1 and 2 years as compared to medical therapy alone.    MOMENTUM 3 trial was a landmark trial for the HM3 device, showing that in a population of end stage HF patients (86% inotrope dependent, 32% INTERMACS 1-2, and 60% DT strategy), 5-year survival with HM3 was 58% and HM3 had lower HRAE's compared with HM2.  There are both patient-specific factors and surgical considerations when it comes to candidacy for LVAD therapy.  RV function prior to LVAD is a key determinant for success post-LVAD  Many patients being considered for LVAD may not have robust RV function, however, predicting RV failure after LVAD is exceedingly difficult.   In general, it doesn’t matter how bad the RV may look on imaging; we care more about the pre-LVAD hemodynamics (look at the PAPi and RA/wedge ratio).   What happens in the OR may be the most important determinant of how the RV will do with the LVAD!  Notes Notes drafted by Dr. Josh Longinow.  1. Historical background of heart pumps and LVADs  LVAD Evolution   FDA approval year  2001  2008  2012  2017  Pump  HeartMate XVE   HeartMate II  Heartware HVAD  HeartMate III  Flow/Design Features  Pulsatile Technology   Continuous flow Axial design  Continuous flow  Centrifugal design  Continuous flow   Full MagLev + Centrifugal design  The 1960's ushered in the first ‘LVADs', when the first air-powered ‘LVAD' was implanted. It kept the patient alive for four days before the patient expired.   The first generation of LVADs were pulsatile pumps   The first nationally recognized, FDA approved LVAD was the HeartMate XVE (late 1990s to early 2000s, REMATCH trial). The XVE pump used compressed air (pneumatically driven) to power the pump.   Prior to the XVE, OHT was the standard of care for patients with advanced, end-stage heart failure.   The second and third generations of LVADs were non-pulsatile, continuous flow devices and included the HVAD, HM2, and HM3 devices.   MOMENTUM 3 was a landmark trial for the HM3 device, showing that in a population of sick patients with end stage HF (86% inotrope dependent, 32% INTERMACS 1-2, and 60% DT strategy), 5-year survival with HM3 was 58% and HM3 had lower HRAE's compared with HM2.   The only pump that is currently FDA approved for implant is the HM3, although other pumps are in clinical trials (BrioVAD system, INNOVATE Trial).  2. What are LVADs, and how do they work?   In simplest terms, the LVAD is a heart pump comprised of several key mechanistic components:   Inflow cannula  Mechanical pump   Outflow cannula  Driveline  Controller/Power source  The HM3 differs from its predecessors (HM2 and HVAD) in several key ways;   HM3 is placed intrapericardial whereas the HM2 was placed pre-peritoneal.   Perhaps most importantly, the HM3 is a fully magnetically levitated, centrifugal flow pump, whereas the HM2 is an axial flow device.  Axial flow pumps are not magnetically levitated, leading to more friction produced between the ruby bearing's contact with the pump rotors, and higher rates of hemocompatibility related adverse events (HRAEs, i.e. pump thrombosis) and the HM2 was ultimately discontinued in favor of the HM3 (MOMENTUM 3 trial).  3. What do the terms ‘Destination Therapy' (DT) or ‘Bridge to Transplant' (BTT) mean when it comes to LVADs?   When LVADs first came on the stage, EVERYONE was a BTT; these early pumps weren't designed for long term use (I.e. REMATCH Trial, Heartmate XVE)  Destination therapy means the LVAD was placed in leu of transplant because there are contraindications to transplant   REMATCH trial brought about the concept of “Destination therapy”, comparing outcomes in patients (with contraindications for transplant) who received an LVAD vs optimal medical therapy  Bridge to transplant means we are placing the LVAD in a patient who may not be a transplant candidate at this moment in time (is too sick, or conversely, not sick enough), but may be down the line   Bridge to recovery is another term used when the LVAD is being placed for a patient we think may have a recoverable cardiomyopathy  4. What are some factors we should consider when assessing a patient’s candidacy for LVAD, in general, and from a surgical perspective?   Patient factors   Older age might push us towards thinking LVAD rather than transplant  In general, age > 70 is the cutoff for transplant, but this is not a hard cut off and varies institution to institution    In general, think about things that help predict recovery after a major surgery; Frailty and Nutritional status are important, we try to optimize these prior to LVAD implant   Right ventricular function remains the Achilles heel of LV support  We know that needing temporary RV support post LVAD puts you on a different survival curve than patients who don’t need RVAD support  Studies have not been able to successfully predict who will develop RV failure after LVAD implantation  What happens in the time between when the patient goes to the OR and when they get back to the ICU is an important determinant who might develop RV failure post LVAD   Surgical techniques such as implanting the HM3 in the intra-thoracic cavity, rather than intra-pericardial may help maintain LV/RV geometry to help optimize the RV post LVAD   Surgical considerations for LVAD candidacy  Small, hypertrophied LV: HM3 inflow cannula is small, but small hypertrophied ventricles tend towards chamber collapse during systole causing suction, needing to run slower with lower flow rates  Chest size/diameter: pumps have gotten so small now, that for adults, these have become less of a consideration  BMI: low BMI used to be more of a concern with the older pumps due to where they were placed, and the relative size of the pump itself, not so much now with the smaller HM 3 pumps  Calcified LV apex: would increase risk of stroke, bleeding   Driveline tunneling becomes a concern in the super obese population, higher risk for driveline infections (might tunnel these driveline's shorter, and to a less fatty region of the abdomen, could even tunnel out the thoracic cavity in the super obese to limit skin motion)    5. Is there a role for MCS (i.e. temporary LVAD such as Impella) in pre-habilitation of patients prior to LVAD surgery?   The theory of being able to improve systemic perfusion, decongest the organs, and make the patient feel better prior to surgery makes sense, but becomes problematic due to the lack of a hard end point/time for prehabilitation which might risk delays in surgery   More likely that it can lead to delay in the surgery, with less-than-optimal benefit; you don't want to prolong the wait for surgery and increase the risk for complications prior to surgery    An Impella 5.5 is currently FDA approved for 2 weeks of support, not 2 months so timing is important to keep in mind  It’s unlikely that you will take a patient and convert them from a malnourished, cachectic person in 2 weeks’ time   6. Is there a role for LVAD therapy in the younger patient population? Should we be thinking of LVAD up front for these patients, with the goal of transplanting down the line?   Recovery may be more likely in certain populations, particularly younger females with smaller LV's; in those populations, perhaps bridge to recovery should be the focus, optimizing them on GDMT etc.   The replacement of transplant, with MCS (LVAD) in young patients has become a topic of discussion, because these pumps have become better and better, with the thinking that an LVAD could bridge a patient for 10 years or so, and they could get a transplant later   It is still a big unknown, but several concerns exist  Patients who get LVADs might end up with complications that become contraindication to transplant down the line (stroke, sensitization etc)   Patients and providers are more hesitant because of the more recent iteration for the UNOS criteria for OHT listing which no longer gives patients with an uncomplicated LVAD higher priority, and therefore they could end up waiting a longer time for a heart after undergoing LVAD  References Rose EA, Gelijns AC, Moskowitz AJ, et al. Long-term use of a left ventricular assist device for end-stage heart failure. N Engl J Med. 2001;345(20):1435-1443. doi:10.1056/NEJMoa012175  Mehra MR, Uriel N, Naka Y, et al. A Fully Magnetically Levitated Left Ventricular Assist Device – Final Report. N Engl J Med. 2019;380(17):1618-1627. doi:10.1056/NEJMoa1900486  Mancini D, Colombo PC. Left Ventricular Assist Devices: A Rapidly Evolving Alternative to Transplant. J Am Coll Cardiol. 2015;65(23):2542-2555. doi:10.1016/j.jacc.2015.04.039  Mehra MR, Goldstein DJ, Cleveland JC, et al. Five-Year Outcomes in Patients With Fully Magnetically Levitated vs Axial-Flow Left Ventricular Assist Devices in the MOMENTUM 3 Randomized Trial. JAMA. 2022;328(12):1233-1242. doi:10.1001/jama.2022.16197  Rose EA, Moskowitz AJ, Packer M, et al. The REMATCH trial: rationale, design, and end points. Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure. Ann Thorac Surg. 1999;67(3):723-730. doi:10.1016/s0003-4975(99)00042-9  Kittleson MM, Shah P, Lala A, et al. INTERMACS profiles and outcomes of ambulatory advanced heart failure patients: A report from the REVIVAL Registry. J Heart Lung Transplant. 2020;39(1):16-26. doi:10.1016/j.healun.2019.08.017  Mehra MR, Netuka I, Uriel N, et al. Aspirin and Hemocompatibility Events With a Left Ventricular Assist Device in Advanced Heart Failure: The ARIES-HM3 Randomized Clinical Trial. JAMA. 2023;330(22):2171-2181. doi:10.1001/jama.2023.23204  Mehra MR, Nayak A, Morris AA, et al. Prediction of Survival After Implantation of a Fully Magnetically Levitated Left Ventricular Assist Device. JACC Heart Fail. 2022;10(12):948-959. doi:10.1016/j.jchf.2022.08.002  Bhardwaj A, Salas de Armas IA, Bergeron A, et al. Prehabilitation Maximizing Functional Mobility in Patients With Cardiogenic Shock Supported on Axillary Impella. ASAIO J. 2024;70(8):661-666. doi:10.1097/MAT.0000000000002170 

La sección 'la hora de los Fósforos' de 'Herrera en COPE', conducido por Alberto Herrera, ha sido el escenario de una historia sorprendente. Un oyente llamado Rafa ha compartido el relato de cómo unos amigos suyos encontraron un tesoro mientras reformaban una antigua casa en Córdoba.Los hechos, que según el narrador ocurrieron hace más de 40 años, tuvieron lugar en una casa muy antigua, de más de cinco siglos, situada en la zona de la Mezquita, por el área de Bataneros. Unos amigos de Rafa compraron la propiedad con la intención de transformarla en un restaurante.Durante las obras, al proceder al derribo de la cocina, que contaba con una vieja estructura de hierro fundido, se toparon con algo inesperado en el tabique que había detrás. Allí encontraron un pañuelo grande que cubría una bolsa llena de joyas y oro antiguo.El descubrimiento se mantuvo en el más estricto secreto para evitar complicaciones sobre la propiedad del tesoro. "Nadie contó allí nada, allí se ...

AUDIO de la charla-coloquio de Dani Zurdo en Bedmar

La política de repliegue promovida por el presidente de Estados Unidos y sus constantes cambios de opinión están empujando a otras potencias a nuevos entendimientos.

Unos cuidados adecuados de la piel tras un tratamiento médico-estético pueden disminuir las molestias, acelerar la recuperación y evitar complicaciones. Por ello, en este episodio, la Dra. Sofía de Asís y la Dra. Inés Escandell seleccionan limpiadores, cremas reparadoras y fotoprotectores recomendables para una rutina cosmética óptima en estos casos, además también darán algún tip de cuidado extra, ¡no te lo pierdas!Más información en nuestra web dermotheque.com y en nuestro perfil de instagram.

¡ La desigualdad en las tareas del hogar puede acabar con tu vida sexual !

AHORA SOMOS UNOS THERIAN - REGRESAMOS 2026

Hermandad, sentimiento y corazón reverenciando a uno de los barrios más musicales de La Habana. Con el título de "Amigos de Santa Amalia", en 1999: el trompetista Julito Padrón, el trombonista Juan Carlos Marín, David Alfaro al piano, David Suárez en saxo, Alfredo Echevarría en bajo, Lukmil Pérez en drums y Alexis Cuesta en las tumbadoras, entre otros, dejaron testimonio de un tiempo único en el devenir de la música popular cubana. Sonidos homenajeando a un barrio lleno de historia donde, en décadas pasadas, convivieron grandes artistas y músicos como los pianistas Robertico Álvarez, Bebo Valdés, Rafael Ortega y Chucho Valdés, el saxofonista René Ravelo, el bongosero Guillermo Romero, y los cantantes Weeno Rodríguez, las hermanas Romay y Mayra Caridad Valdés, figuras que durante años mantuvieron vivo el pulso artístico en sus calles. Le pusieron voz y sentimiento a "Los amigos de Santa Amalia": Julito Padrón, Aramis Galindo y Moraima Marín. Amado de Jesús Dedeu siempre será recordado entonando sereno, clave en mano, los aires de la rumba y el guaguancó. Unos minutos para recordar al veterano rumbero que le entregó alma, corazón y vida al canto y los toques que, en esencia, continúan definiendo el mapa sonoro de Cuba. Falleció en La Habana el 14 de febrero de 2026. En la memoria el maestro Dedeu con su Grupo "Clave y Guaguancó". Boris Larramendi compuso los temas que hoy completan su más reciente álbum, ubicado ya en todas las plataformas digitales, entre 2021 y 2025. Su título: "Oye". Fiel en esencia al sonido que ha caracterizado su permanencia en los escenarios por más de treinta años, para completar esta nueva aventura discográfica, contó con viejos colegas: Nam San Fong en la co-producción y las guitarras eléctricas, Johan Medina en la batería, Eduardo Rodríguez en la percusión, Daniel Stable en el bajo y Dianela de la Portilla en los coros. Como invitada especial: Ivette Falcón en violoncello y los infaltables "Habana Abierta": Vanito, Alejandro, Medina, Kelvis y Barbería. En la mezcla y el mastering de este nuevo trabajo de Boris: Oscar Autie, en el Cerrito Records. En el diseño de la portada que hoy identifica nuestro programa: la talentosa artista plástica Camila Lobón. Un lujo compartir contigo estos sonidos en primicia, invitándote a encontrarlo en las plataformas digitales. Los años 40 del siglo XX lo vieron despuntar como uno de los pianistas, compositores y arreglistas fundamentales en el mismo centro de la llamada "era de los conjuntos soneros". Sin embargo el quehacer de Luis Martinez Griñán, ya entonces bien conocido en el ambiente musical como "Lilí", inmerso en el patrón estilístico establecido por el tresero Arsenio Rodríguez, tendrá mayor campo de expresión en la década siguiente cuando Chappotin se hace cargo del conjunto del ciego maravilloso. Sus arreglos donde los riffs de los metales se abrieron más al swing, así como las cadencias de los montunos que, aún marcadas por la clave, resultaron menos recias, le permitieron perfilar su progresivo y original concepto interpretativo. La memoria nos trae de vuelta algunas piezas del gran Lilí Martinez, el músico innovador, el pianista sonero cuyas grabaciones, a pesar de los años, mantienen intactas frescura, sabor y, por supuesto, muchísima cubanía. "La perla de Oriente" vió la luz en Guantanamo el 19 de agosto de 1915. Muy próximo el centenario de su natalicio es sumamente importante tener en cuenta su legado.

Previa del FC Barcelona-Levante que puede dejar a los azulgranas como nuevo líder de LaLiga. Además, lo mejor del fútbol internacional y de la Segunda División. Entrevista con Oriol Cardona tras una nueva medalla en los Juegos de Invierno para España y, por último, charla con Antonio de la Rosa. 

Previa del FC Barcelona-Levante que puede dejar a los azulgranas como nuevo líder de LaLiga. Además, lo mejor del fútbol internacional y de la Segunda División. Entrevista con Oriol Cardona tras una nueva medalla en los Juegos de Invierno para España y, por último, charla con Antonio de la Rosa. 

Wendy, de Nueva York

En la Encuesta Absurda de hoy, Fernando Martín, ha llamado a Gracia y le ha preguntado cosas como que: ¿si una presentadora de televisión tiene problemas estomacales es "La Revuelta"?; ¿Trabajar en una empresa de zapatos ortopédicos significa "estar en plantilla"? Escucha de lunes a jueves, en CADENA 100, la Encuesta Absurda de Fernando Martín.

DISPONIBLE EN YOUTUBE

Terminar con el racismo y la xenofobia no es algo que se consiga de la noche a la mañana. De hecho, es prácticamente imposible que se erradiquen completamente. Hay seres mononeuronales inasequibles al desaliento y que son irrecuperables. De la misma forma, hay que tener cuidado con los aliados coyunturales, esos que se suman hoy porque convuene, pero que mañana te pegarán la puñalada trapera por la espalda. Min. 01 Seg. 53 – Intro Min. 07 Seg. 57 - Un problema que hay que cortar de raíz Min. 17 Seg. 30 - Si te sancionan, te hacen más grande Min. 26 Seg. 51 – Unos aliados poco fiables Min. 35 Seg. 19 - Hay que dar un paso más allá Min. 42 Seg. 42 - La otra versión de la historia Min. 48 Seg. 31 - Precedentes que, supuestamente, nos hicieron mejores Min. 55 Seg. 34 - Empatía de casi todo el mundo Min. 60 Seg. 27 - Un baile poco edificante Min. 65 Seg. 24 - Un partido industrial y un gran problema Min. 71 Seg. 09 - Despedida Crosby, Stills Nash & Young (Chicago, IL 27/08/1974) Helpless Almost Cut My Hair Wooden Ships Teach Your Cjildren Johnny's Garden Love The One You're With The Lee Shore Star Of Bethlehem Southbound Train Immigration Man Our House Molly Tuttle & Golden Highway - Crooked Tree (Nashville, TN 28/03/2022)

- En el portal de Carmen Aristegui quedó alojada desde ayer una bomba que confirma la podredumbre del Narco Régimen Morenista, al más alto nivel. - Se trata de un audio de 20 minutos en el que dos secretarios de marina, uno entrante y otro saliente, reciben la denuncia directa de un integrante de la Marina, sobre el caso de corrupción más grande de la historia de México, que es el del huachicol fiscal, y le piden que lo ponga por escrito. - Unos días después, el denunciante entrega una carta describiendo lo que narró en la reunión. - Unos días después de entregar la carta de denuncia, es asesinado. - ¿Qué más necesitas saber para convencerte de la podredumbre del Narco Régimen Morenista? Hosted by Simplecast, an AdsWizz company. See https://pcm.adswizz.com for information about our collection and use of personal data for advertising.

El análisis de actualidad económica, todas las mañanas a las 07:30. 

Trump dijo que en unos 10 días se sabrá si EE.UU. logra un acuerdo con Irán sobre su programa nuclear o si considera una acción militar.

- En el portal de Carmen Aristegui quedó alojada desde ayer una bomba que confirma la podredumbre del Narco Régimen Morenista, al más alto nivel. - Se trata de un audio de 20 minutos en el que dos secretarios de marina, uno entrante y otro saliente, reciben la denuncia directa de un integrante de la Marina, sobre el caso de corrupción más grande de la historia de México, que es el del huachicol fiscal, y le piden que lo ponga por escrito. - Unos días después, el denunciante entrega una carta describiendo lo que narró en la reunión. - Unos días después de entregar la carta de denuncia, es asesinado. - ¿Qué más necesitas saber para convencerte de la podredumbre del Narco Régimen Morenista? Hosted by Simplecast, an AdsWizz company. See https://pcm.adswizz.com for information about our collection and use of personal data for advertising.

Escuche esta y más noticias de LA PATRIA Radio de lunes a viernes por los 1540 AM de Radio Cóndor en Manizales y en www.lapatria.com, encuentre videos de las transmisiones en nuestro Facebook Live: www.facebook.com/lapatria.manizales/videos

Enamorados en la escuela las cosas cambiaron rápidamenteECDQEMSD podcast episodio 6242 Qué pasó con el Amor?Conducen: El Pirata y El Sr. Lagartija https://canaltrans.comNoticias Del Mundo: El Kilauea y el fuego - Paro anunciado - Centrales Nucleares por Europa - Operativo en El Salvador - La junta de paz - Señora Trevi siéntese - Playboy inteligente - Los burdeles de Valencia - Cambia la alerta sísmica.Historias Desintegradas: El chico que me deslumbró - Incomoda situación - Unos tragos y unos besos - Enamoradísima - Qué somos? - El derrumbe - Échale salsa - Clase de tamales yucatecos - Los permitidos - La falsa perla - Jorge sea como sea - El gato italiano - Kosovo - Juego responsable - Inventores mexicanos y más...En Caso De Que El Mundo Se Desintegre - Podcast no tiene publicidad, sponsors ni organizaciones que aporten para mantenerlo al aire. Solo el sistema cooperativo de los que aportan a través de las suscripciones hacen posible que todo esto siga siendo una realidad. Gracias Dragones Dorados!!NO AI: ECDQEMSD Podcast no utiliza ninguna inteligencia artificial de manera directa para su realización. Diseño, guionado, música, edición y voces son de  nuestra completa intervención humana.

CADENA 100 presenta la Encuesta Absurda en '¡Buenos días, Javi y Mar!'. Christian ha respondido a preguntas  como qué opina de niños que juegan al fútbol con un Roomba o si la palabra "velcro" se escribe como suena. También habla sobre si un portero de fútbol es hiperactivo por no parar. Finalmente, se preguntan si las críticas de un albañil son constructivas. ¡Escucha de lunes a jueves la Encuesta Absurda de Fernando Martín!

Japón celebró hace unos días elecciones anticipadas. Unos comiciosque han confirmado la popularidad del actual primera ministra nipona Sanae Takahichi. Tiene un perfil marcadamente nacionalista, ha mostrado su preocupación por las pretensiones chinas sobre Taiwán y ha apostado por fortalecer militarmente a su país y profundizar en su alianza económica pero también a nivel de seguridad con Estados Unidos. Hablamos con Georgina Higueras, directora de #ForoAsia en la Fundación Foro de Foros y periodista especializada en el continente asiático.Escuchar audio

01 16-02-26 LHDW Los políticos no son buenas personas, además de ser unos golfos. Óscar López atacando al difunto Lambán por los malos resultados en Aragón del PSOE

Juan Leon tiene un negocio de grúas que se llama Leo's Towing en las Ciudades Gemelas. Unos meses después de empezar, vio algo que lo puso triste. – en las calles y en los lugares de aparcar por muchos días. Los dueños no estaban porque se los había llevado la policía de inmigración. “Vi que El conductor de grúa ayuda a familias a recuperar sus coches Read More » Read the full Article: El conductor de grúa ayuda a familias a recuperar sus coches

Esther Borja continúa siendo una de las figuras representativas del canto lírico cubano. A pesar del vastisimo repertorio de compositores que alcanzó relevancia en su voz, las obras del maestro Ernesto Lecuona tuvieron una marcada presencia. "Cuba", hermosa pieza del genio de Guanabacoa que en una presentación radiofónica le acompañó al piano, nos permitió comenzar este breve segmento. En un ambiente teatral marcado por el reinado de la zarzuela, el género español adaptado a la criolla por numerosos libretistas y compositores donde resaltó Lecuona, Esther Borja nació en los escenarios. Fue tal el exito del vals canción "Damisela encantadora" que, durante más de 60 años se hizo indispensable en su repertorio. El antológico catálogo discográfico de la etiqueta norteamericana Víctor de finales de los años 30 del siglo XX, editado en placas de 78 rpm, nos trae en voz de Esther una olvidada canción de Bola de Nieve: "Arroyito de mi casa". Le acompañó la orquesta del maestro Alfredo Brito. Siempre Lecuona en la voz de Esther Borja. Durante la década del 40 y comienzos de los 50, etapa de intensa actividad teatral, junto a otras importantes voces del canto lírico, fue parte de aquellas compañías teatrales que, bajo la batuta del maestro, difundieron incesantemente la canción cubana por el mundo. Buena parte de ese bagaje cultivado al lado de otras notables figuras del canto de su tiempo como Hortensia Coalla, Zoraida Marrero, Panchito Naya, Manolo Álvarez Mera, María de los Angeles Santana, Sarita Escarpenter, entre muchas otras, lo ofreció Esther gustosamente a su público desde "Álbum de Cuba", el programa que condujo magistralmente durante 25 años en la televisión. Esther Borja en la memoria. Murió centenaria, el 28 de diciembre del 2013. Había nacido en La Habana el 5 de diciembre de 1913. Un recuerdo para el gran Alden Knight. Talentoso y carismático actor se ganó un sitial importante en el firmamento de los grandes de Cuba. Artista todo terreno jamás existieron fronteras para que ofreciera su arte en todos los escenarios posibles. Así, durante más de setenta años, brilló con luz propia en el teatro, el cabaret, la radio, el cine y la televisión. Actor, declamador, presentador, bailarín, locutor, productor y cantante, dejó su huella en más de una generación de cubanos. Falleció en La Habana el 11 de febrero de 2026 a los 89 años. Hoy el más tierno y entrañable recuerdo nos devuelve a Alden Knight interpretando canciones infantiles. Regresamos a la discografia independiente cubana para resaltar la labor musical del recientemente desaparecido maestro Roberto Sánchez Ferrer. El 4 de febrero de 2026 falleció a los 99 años este notable músico. Saxofonista y clarinetista, ocupó atriles en importantes formaciones como las jazz bands "Hermanos Castro", "Havana Casino", "Hermanos LeBatard" y "Riverside". Su refinado concepto melódico y armónico impulsó su formación como arreglista, labor que, a partir de los años 50, desarrolló con mayor profusión ganando prestigio entre sus contemporáneos, definiendo con sus trabajos para cuerdas y jazz bands un sonido característico, de honda cubanía, vinculando lo popular y lo sinfónico. Unos minutos para que nos acompañe el maestro Roberto Sánchez Ferrer con algo de su legado discográfico. Del álbum "Fantasía cubana", producido por la etiqueta Kubaney durante la segunda mitad de los 50s, escogemos los clásicos: "Almendra" de Abelardito Valdés y "Son de la loma" de Miguel Matamoros. De "Canciones cubanas", álbum producido por la etiqueta Puchito al tenor Manolo Álvarez Mera: la habanera "Tú" de Eduardo Sánchez de Fuentes y, desde una producción Gema, Fernando Álvarez cantará el bolero de Ela O'Farrill: "No tienes por qué criticar". Sentimiento curado con el poderío y el sabor de ancestrales cantos y toques de tambor. Infinitas gracias, una vez más, a los hermanos Abreu. Fiesta en grande despedirnos con Los Papines y la inmensa Celeste Mendoza.

«Eran inocentes porque eran chicos.... »Corrían, jugaban, y sus risas eran inconscientes vibraciones de vida en los jardines.... Sentábanse... sobre el rústico banco de la glorieta, y él contaba historias que le habían leído, mientras jugaba con los deditos de su compañera atenta. »Eran cuentos como todos los juegos infantiles, en que sucedían cosas fantásticas, en que había príncipes y princesitas que se amaban desesperadamente al través de un impedimento, hasta el episodio final, producido a tiempo para hacerlos felices, felices en un amor sin contrariedades.... »Ya tenía él el orgullo viril de ver colgada de sus palabras la atención de esa mujercita, digna de todos los altares. Y cuando su voz se empañaba de emoción al finalizar un cuento, se estrechaban cerca, muy cerca, en busca de felicidad.... »Estaban un día ajenos a todo. El cuento de la princesa rubia había puesto entre ellos la ascendencia de su fantasía. Ella se arrebujaba contra él desparramando en hilachas de oro sus bucles sobre el hombro amigo; él la había atraído lo más posible y besaba, como estampas sagradas, sus ojos, trémulos de promesas ignotas.»1 Así nos describe Ricardo Güiraldes, en su cuento titulado «Sexto», el primer amor con el que los más jóvenes sueñan y los menos jóvenes se identifican. ¡Qué bien logradas esas imágenes del muchacho que le cuenta historias a su atenta compañera «colgada de sus palabras» mientras juega con sus delicados dedos, y de «esa mujercita, digna de todos los altares», cuyos ojos él besa «como estampas sagradas»! No persiguen más que lo que parecen encontrar los protagonistas de sus cuentos fantásticos: el ser «felices en un amor sin contrariedades». Este es uno de una colección de cuentos que Güiraldes comenzó a escribir en su adolescencia, pero terminó en París, lejos de su patria argentina, entre 1911 y 1912.2 Unos mil ochocientos años antes, el apóstol Juan había abordado el mismo tema del primer amor al escribirle a la Iglesia de Éfeso, desde donde había sido desterrado a la isla de Patmos. Allí, en el Apocalipsis, le escribió: «Tengo en tu contra que has abandonado tu primer amor».3 Sin embargo, a diferencia de Güiraldes, el primer amor al que se refería San Juan no era físico sino espiritual. Era el amor que al principio los efesios le habían manifestado a su Señor y Salvador Jesucristo. Al primer amor físico sólo podemos volver mediante remembranzas del ayer como las que evoca Güiraldes, porque en lo físico las dos partes han cambiado para siempre. En cambio, al primer amor espiritual sí podemos volver porque una de las dos partes, Dios, no ha cambiado en absoluto4 desde que primero lo amamos. Así como los efesios, sólo tenemos que arrepentirnos y amarlo como al principio.5 Dios nos espera con brazos abiertos, y quiere rodearnos estrechamente con los lazos de su amor eterno.6 Carlos ReyUn Mensaje a la Concienciawww.conciencia.net 1 Ricardo Güiraldes, Cuentos de muerte y de sangre (Buenos Aires: Editorial Losada, 1978), pp. 111-112. 2 Ibíd., p. 11. 3 Ap 2:4 4 Stg 1:17 5 Ap 2:5 6 Jer 31:3

En "Una advertencia a Miss Universo" investigan quién ha enviado una nota amenazante a las candidatas. El Club de los Viudos Negros', de Asimov. CENA DE FEBRERO 🍷🍰 📍 Ristorante Casa Milano – Milano, Italia 🧭 Coordenadas: 45°28'19.8"N 9°12'06.4"E Isaac Asimov los creó como un homenaje al placer de conversar, al arte de observar y a la deliciosa costumbre de no quedarse con la primera respuesta. Acomódate. El vino está servido. La cena va a comenzar. Y tú… Tú también estás invitado. Un círculo discreto de seis caballeros que se reúnen una vez al mes, siempre en el mismo restaurante, siempre en la misma mesa, y siempre con una única regla: cada cena debe tener un invitado, y ese invitado debe estar dispuesto a hablar y a ser interrogado. 🕷🕷🕷🕷🕷🕷🕸 Los Viudos Negros son un club de seis hombres que se reúnen una vez al mes en un reservado del restaurante Milano de Nueva York. Cada noche uno de ellos preside el encuentro y tiene el derecho de llevar un invitado, al que interrogan. Al principio sólo se reunían para comer y conversar pero últimamente uno de ellos plantea algún tipo de problema o delito. Los miembros del club buscan respuestas complejas a los enigmas planteados y luego Henry, el camarero, descubre la simple verdad. El club está formado por:🍷🍷🍷🍷🍷🍷 Geoffrey Avalon, Jeff. Alto y delgado, espesas cejas negras, bigote recortado y barbita gris. Fue oficial durante la II Guerra Mundial y trabaja como abogado en derecho patentario. Mario Gonzalo, pintor y gran artista. Thomas Trumbull. Rostro moreno y arrugado, permanentemente descontento. Experto en códigos, alto consejero del gobierno. Emmanuel Rubin, Manny. Bajito, mide 1,55, barba rala, lentes gruesos. Fue predicador adventista con 15 años y conoce bien la Biblia. Está casado y es escritor de novelas policíacas. James Drake. Bigote. Vive en New Jersey. Especialista en química orgánica con amplios conocimientos en literatura. Roger Halsted, calvo. Profesor de matemáticas en una escuela secundaria. Escribe la Ilíada en quintillas y todos los meses les recita una estrofa. Es miembro de los Irregulares de Baker Street. Henry Jackson, el camarero. Unos 60 años, sin arrugas. Es humilde y honrado. Entre ellos se llaman doctores y si uno es doctor de carrera le denominan doctor doctor. Para ayudarse en sus investigaciones cuentan con diccionarios, biblias y las obras de Shakespeare en su biblioteca. Y recuerda que puedes seguirnos en Telegram, YouTube, Instagram y X, y si este podcast te acompaña, te inspira o te gusta lo que hago, puedes hacerte fan y apoyar la nave. Tu energía mantiene viva esta aventura sonora.🚀 Aquí te dejo la página directa para apoyarme: 🍻 https://www.ivoox.com/support/552842 ¡¡Muchas gracias por todos tus comentarios y por tu apoyo!! Voz y sonido Olga Paraíso, Música epidemic sound con licencia premium autorizada para este podcast. ⏩BIO Olga Paraíso: https://instabio.cc/Hleidas PLAYLIST EL CLUB DE LOS VIUDOS NEGROS EN Ivoox https://go.ivoox.com/bk/11290149 Escucha el episodio completo en la app de iVoox, o descubre todo el catálogo de iVoox Originals

From rewriting Google's search stack in the early 2000s to reviving sparse trillion-parameter models and co-designing TPUs with frontier ML research, Jeff Dean has quietly shaped nearly every layer of the modern AI stack. As Chief AI Scientist at Google and a driving force behind Gemini, Jeff has lived through multiple scaling revolutions from CPUs and sharded indices to multimodal models that reason across text, video, and code.Jeff joins us to unpack what it really means to “own the Pareto frontier,” why distillation is the engine behind every Flash model breakthrough, how energy (in picojoules) not FLOPs is becoming the true bottleneck, what it was like leading the charge to unify all of Google's AI teams, and why the next leap won't come from bigger context windows alone, but from systems that give the illusion of attending to trillions of tokens.We discuss:* Jeff's early neural net thesis in 1990: parallel training before it was cool, why he believed scaling would win decades early, and the “bigger model, more data, better results” mantra that held for 15 years* The evolution of Google Search: sharding, moving the entire index into memory in 2001, softening query semantics pre-LLMs, and why retrieval pipelines already resemble modern LLM systems* Pareto frontier strategy: why you need both frontier “Pro” models and low-latency “Flash” models, and how distillation lets smaller models surpass prior generations* Distillation deep dive: ensembles → compression → logits as soft supervision, and why you need the biggest model to make the smallest one good* Latency as a first-class objective: why 10–50x lower latency changes UX entirely, and how future reasoning workloads will demand 10,000 tokens/sec* Energy-based thinking: picojoules per bit, why moving data costs 1000x more than a multiply, batching through the lens of energy, and speculative decoding as amortization* TPU co-design: predicting ML workloads 2–6 years out, speculative hardware features, precision reduction, sparsity, and the constant feedback loop between model architecture and silicon* Sparse models and “outrageously large” networks: trillions of parameters with 1–5% activation, and why sparsity was always the right abstraction* Unified vs. specialized models: abandoning symbolic systems, why general multimodal models tend to dominate vertical silos, and when vertical fine-tuning still makes sense* Long context and the illusion of scale: beyond needle-in-a-haystack benchmarks toward systems that narrow trillions of tokens to 117 relevant documents* Personalized AI: attending to your emails, photos, and documents (with permission), and why retrieval + reasoning will unlock deeply personal assistants* Coding agents: 50 AI interns, crisp specifications as a new core skill, and how ultra-low latency will reshape human–agent collaboration* Why ideas still matter: transformers, sparsity, RL, hardware, systems — scaling wasn't blind; the pieces had to multiply togetherShow Notes:* Gemma 3 Paper* Gemma 3* Gemini 2.5 Report* Jeff Dean's “Software Engineering Advice fromBuilding Large-Scale Distributed Systems” Presentation (with Back of the Envelope Calculations)* Latency Numbers Every Programmer Should Know by Jeff Dean* The Jeff Dean Facts* Jeff Dean Google Bio* Jeff Dean on “Important AI Trends” @Stanford AI Club* Jeff Dean & Noam Shazeer — 25 years at Google (Dwarkesh)—Jeff Dean* LinkedIn: https://www.linkedin.com/in/jeff-dean-8b212555* X: https://x.com/jeffdeanGoogle* https://google.com* https://deepmind.googleFull Video EpisodeTimestamps00:00:04 — Introduction: Alessio & Swyx welcome Jeff Dean, chief AI scientist at Google, to the Latent Space podcast00:00:30 — Owning the Pareto Frontier & balancing frontier vs low-latency models00:01:31 — Frontier models vs Flash models + role of distillation00:03:52 — History of distillation and its original motivation00:05:09 — Distillation's role in modern model scaling00:07:02 — Model hierarchy (Flash, Pro, Ultra) and distillation sources00:07:46 — Flash model economics & wide deployment00:08:10 — Latency importance for complex tasks00:09:19 — Saturation of some tasks and future frontier tasks00:11:26 — On benchmarks, public vs internal00:12:53 — Example long-context benchmarks & limitations00:15:01 — Long-context goals: attending to trillions of tokens00:16:26 — Realistic use cases beyond pure language00:18:04 — Multimodal reasoning and non-text modalities00:19:05 — Importance of vision & motion modalities00:20:11 — Video understanding example (extracting structured info)00:20:47 — Search ranking analogy for LLM retrieval00:23:08 — LLM representations vs keyword search00:24:06 — Early Google search evolution & in-memory index00:26:47 — Design principles for scalable systems00:28:55 — Real-time index updates & recrawl strategies00:30:06 — Classic “Latency numbers every programmer should know”00:32:09 — Cost of memory vs compute and energy emphasis00:34:33 — TPUs & hardware trade-offs for serving models00:35:57 — TPU design decisions & co-design with ML00:38:06 — Adapting model architecture to hardware00:39:50 — Alternatives: energy-based models, speculative decoding00:42:21 — Open research directions: complex workflows, RL00:44:56 — Non-verifiable RL domains & model evaluation00:46:13 — Transition away from symbolic systems toward unified LLMs00:47:59 — Unified models vs specialized ones00:50:38 — Knowledge vs reasoning & retrieval + reasoning00:52:24 — Vertical model specialization & modules00:55:21 — Token count considerations for vertical domains00:56:09 — Low resource languages & contextual learning00:59:22 — Origins: Dean's early neural network work01:10:07 — AI for coding & human–model interaction styles01:15:52 — Importance of crisp specification for coding agents01:19:23 — Prediction: personalized models & state retrieval01:22:36 — Token-per-second targets (10k+) and reasoning throughput01:23:20 — Episode conclusion and thanksTranscriptAlessio Fanelli [00:00:04]: Hey everyone, welcome to the Latent Space podcast. This is Alessio, founder of Kernel Labs, and I'm joined by Swyx, editor of Latent Space. Shawn Wang [00:00:11]: Hello, hello. We're here in the studio with Jeff Dean, chief AI scientist at Google. Welcome. Thanks for having me. It's a bit surreal to have you in the studio. I've watched so many of your talks, and obviously your career has been super legendary. So, I mean, congrats. I think the first thing must be said, congrats on owning the Pareto Frontier.Jeff Dean [00:00:30]: Thank you, thank you. Pareto Frontiers are good. It's good to be out there.Shawn Wang [00:00:34]: Yeah, I mean, I think it's a combination of both. You have to own the Pareto Frontier. You have to have like frontier capability, but also efficiency, and then offer that range of models that people like to use. And, you know, some part of this was started because of your hardware work. Some part of that is your model work, and I'm sure there's lots of secret sauce that you guys have worked on cumulatively. But, like, it's really impressive to see it all come together in, like, this slittily advanced.Jeff Dean [00:01:04]: Yeah, yeah. I mean, I think, as you say, it's not just one thing. It's like a whole bunch of things up and down the stack. And, you know, all of those really combine to help make UNOS able to make highly capable large models, as well as, you know, software techniques to get those large model capabilities into much smaller, lighter weight models that are, you know, much more cost effective and lower latency, but still, you know, quite capable for their size. Yeah.Alessio Fanelli [00:01:31]: How much pressure do you have on, like, having the lower bound of the Pareto Frontier, too? I think, like, the new labs are always trying to push the top performance frontier because they need to raise more money and all of that. And you guys have billions of users. And I think initially when you worked on the CPU, you were thinking about, you know, if everybody that used Google, we use the voice model for, like, three minutes a day, they were like, you need to double your CPU number. Like, what's that discussion today at Google? Like, how do you prioritize frontier versus, like, we have to do this? How do we actually need to deploy it if we build it?Jeff Dean [00:02:03]: Yeah, I mean, I think we always want to have models that are at the frontier or pushing the frontier because I think that's where you see what capabilities now exist that didn't exist at the sort of slightly less capable last year's version or last six months ago version. At the same time, you know, we know those are going to be really useful for a bunch of use cases, but they're going to be a bit slower and a bit more expensive than people might like for a bunch of other broader models. So I think what we want to do is always have kind of a highly capable sort of affordable model that enables a whole bunch of, you know, lower latency use cases. People can use them for agentic coding much more readily and then have the high-end, you know, frontier model that is really useful for, you know, deep reasoning, you know, solving really complicated math problems, those kinds of things. And it's not that. One or the other is useful. They're both useful. So I think we'd like to do both. And also, you know, through distillation, which is a key technique for making the smaller models more capable, you know, you have to have the frontier model in order to then distill it into your smaller model. So it's not like an either or choice. You sort of need that in order to actually get a highly capable, more modest size model. Yeah.Alessio Fanelli [00:03:24]: I mean, you and Jeffrey came up with the solution in 2014.Jeff Dean [00:03:28]: Don't forget, L'Oreal Vinyls as well. Yeah, yeah.Alessio Fanelli [00:03:30]: A long time ago. But like, I'm curious how you think about the cycle of these ideas, even like, you know, sparse models and, you know, how do you reevaluate them? How do you think about in the next generation of model, what is worth revisiting? Like, yeah, they're just kind of like, you know, you worked on so many ideas that end up being influential, but like in the moment, they might not feel that way necessarily. Yeah.Jeff Dean [00:03:52]: I mean, I think distillation was originally motivated because we were seeing that we had a very large image data set at the time, you know, 300 million images that we could train on. And we were seeing that if you create specialists for different subsets of those image categories, you know, this one's going to be really good at sort of mammals, and this one's going to be really good at sort of indoor room scenes or whatever, and you can cluster those categories and train on an enriched stream of data after you do pre-training on a much broader set of images. You get much better performance. If you then treat that whole set of maybe 50 models you've trained as a large ensemble, but that's not a very practical thing to serve, right? So distillation really came about from the idea of, okay, what if we want to actually serve that and train all these independent sort of expert models and then squish it into something that actually fits in a form factor that you can actually serve? And that's, you know, not that different from what we're doing today. You know, often today we're instead of having an ensemble of 50 models. We're having a much larger scale model that we then distill into a much smaller scale model.Shawn Wang [00:05:09]: Yeah. A part of me also wonders if distillation also has a story with the RL revolution. So let me maybe try to articulate what I mean by that, which is you can, RL basically spikes models in a certain part of the distribution. And then you have to sort of, well, you can spike models, but usually sometimes... It might be lossy in other areas and it's kind of like an uneven technique, but you can probably distill it back and you can, I think that the sort of general dream is to be able to advance capabilities without regressing on anything else. And I think like that, that whole capability merging without loss, I feel like it's like, you know, some part of that should be a distillation process, but I can't quite articulate it. I haven't seen much papers about it.Jeff Dean [00:06:01]: Yeah, I mean, I tend to think of one of the key advantages of distillation is that you can have a much smaller model and you can have a very large, you know, training data set and you can get utility out of making many passes over that data set because you're now getting the logits from the much larger model in order to sort of coax the right behavior out of the smaller model that you wouldn't otherwise get with just the hard labels. And so, you know, I think that's what we've observed. Is you can get, you know, very close to your largest model performance with distillation approaches. And that seems to be, you know, a nice sweet spot for a lot of people because it enables us to kind of, for multiple Gemini generations now, we've been able to make the sort of flash version of the next generation as good or even substantially better than the previous generations pro. And I think we're going to keep trying to do that because that seems like a good trend to follow.Shawn Wang [00:07:02]: So, Dara asked, so it was the original map was Flash Pro and Ultra. Are you just sitting on Ultra and distilling from that? Is that like the mother load?Jeff Dean [00:07:12]: I mean, we have a lot of different kinds of models. Some are internal ones that are not necessarily meant to be released or served. Some are, you know, our pro scale model and we can distill from that as well into our Flash scale model. So I think, you know, it's an important set of capabilities to have and also inference time scaling. It can also be a useful thing to improve the capabilities of the model.Shawn Wang [00:07:35]: And yeah, yeah, cool. Yeah. And obviously, I think the economy of Flash is what led to the total dominance. I think the latest number is like 50 trillion tokens. I don't know. I mean, obviously, it's changing every day.Jeff Dean [00:07:46]: Yeah, yeah. But, you know, by market share, hopefully up.Shawn Wang [00:07:50]: No, I mean, there's no I mean, there's just the economics wise, like because Flash is so economical, like you can use it for everything. Like it's in Gmail now. It's in YouTube. Like it's yeah. It's in everything.Jeff Dean [00:08:02]: We're using it more in our search products of various AI mode reviews.Shawn Wang [00:08:05]: Oh, my God. Flash past the AI mode. Oh, my God. Yeah, that's yeah, I didn't even think about that.Jeff Dean [00:08:10]: I mean, I think one of the things that is quite nice about the Flash model is not only is it more affordable, it's also a lower latency. And I think latency is actually a pretty important characteristic for these models because we're going to want models to do much more complicated things that are going to involve, you know, generating many more tokens from when you ask the model to do so. So, you know, if you're going to ask the model to do something until it actually finishes what you ask it to do, because you're going to ask now, not just write me a for loop, but like write me a whole software package to do X or Y or Z. And so having low latency systems that can do that seems really important. And Flash is one direction, one way of doing that. You know, obviously our hardware platforms enable a bunch of interesting aspects of our, you know, serving stack as well, like TPUs, the interconnect between. Chips on the TPUs is actually quite, quite high performance and quite amenable to, for example, long context kind of attention operations, you know, having sparse models with lots of experts. These kinds of things really, really matter a lot in terms of how do you make them servable at scale.Alessio Fanelli [00:09:19]: Yeah. Does it feel like there's some breaking point for like the proto Flash distillation, kind of like one generation delayed? I almost think about almost like the capability as a. In certain tasks, like the pro model today is a saturated, some sort of task. So next generation, that same task will be saturated at the Flash price point. And I think for most of the things that people use models for at some point, the Flash model in two generation will be able to do basically everything. And how do you make it economical to like keep pushing the pro frontier when a lot of the population will be okay with the Flash model? I'm curious how you think about that.Jeff Dean [00:09:59]: I mean, I think that's true. If your distribution of what people are asking people, the models to do is stationary, right? But I think what often happens is as the models become more capable, people ask them to do more, right? So, I mean, I think this happens in my own usage. Like I used to try our models a year ago for some sort of coding task, and it was okay at some simpler things, but wouldn't do work very well for more complicated things. And since then, we've improved dramatically on the more complicated coding tasks. And now I'll ask it to do much more complicated things. And I think that's true, not just of coding, but of, you know, now, you know, can you analyze all the, you know, renewable energy deployments in the world and give me a report on solar panel deployment or whatever. That's a very complicated, you know, more complicated task than people would have asked a year ago. And so you are going to want more capable models to push the frontier in the absence of what people ask the models to do. And that also then gives us. Insight into, okay, where does the, where do things break down? How can we improve the model in these, these particular areas, uh, in order to sort of, um, make the next generation even better.Alessio Fanelli [00:11:11]: Yeah. Are there any benchmarks or like test sets they use internally? Because it's almost like the same benchmarks get reported every time. And it's like, all right, it's like 99 instead of 97. Like, how do you have to keep pushing the team internally to it? Or like, this is what we're building towards. Yeah.Jeff Dean [00:11:26]: I mean, I think. Benchmarks, particularly external ones that are publicly available. Have their utility, but they often kind of have a lifespan of utility where they're introduced and maybe they're quite hard for current models. You know, I, I like to think of the best kinds of benchmarks are ones where the initial scores are like 10 to 20 or 30%, maybe, but not higher. And then you can sort of work on improving that capability for, uh, whatever it is, the benchmark is trying to assess and get it up to like 80, 90%, whatever. I, I think once it hits kind of 95% or something, you get very diminishing returns from really focusing on that benchmark, cuz it's sort of, it's either the case that you've now achieved that capability, or there's also the issue of leakage in public data or very related kind of data being, being in your training data. Um, so we have a bunch of held out internal benchmarks that we really look at where we know that wasn't represented in the training data at all. There are capabilities that we want the model to have. Um, yeah. Yeah. Um, that it doesn't have now, and then we can work on, you know, assessing, you know, how do we make the model better at these kinds of things? Is it, we need different kind of data to train on that's more specialized for this particular kind of task. Do we need, um, you know, a bunch of, uh, you know, architectural improvements or some sort of, uh, model capability improvements, you know, what would help make that better?Shawn Wang [00:12:53]: Is there, is there such an example that you, uh, a benchmark inspired in architectural improvement? Like, uh, I'm just kind of. Jumping on that because you just.Jeff Dean [00:13:02]: Uh, I mean, I think some of the long context capability of the, of the Gemini models that came, I guess, first in 1.5 really were about looking at, okay, we want to have, um, you know,Shawn Wang [00:13:15]: immediately everyone jumped to like completely green charts of like, everyone had, I was like, how did everyone crack this at the same time? Right. Yeah. Yeah.Jeff Dean [00:13:23]: I mean, I think, um, and once you're set, I mean, as you say that needed single needle and a half. Hey, stack benchmark is really saturated for at least context links up to 1, 2 and K or something. Don't actually have, you know, much larger than 1, 2 and 8 K these days or two or something. We're trying to push the frontier of 1 million or 2 million context, which is good because I think there are a lot of use cases where. Yeah. You know, putting a thousand pages of text or putting, you know, multiple hour long videos and the context and then actually being able to make use of that as useful. Try to, to explore the über graduation are fairly large. But the single needle in a haystack benchmark is sort of saturated. So you really want more complicated, sort of multi-needle or more realistic, take all this content and produce this kind of answer from a long context that sort of better assesses what it is people really want to do with long context. Which is not just, you know, can you tell me the product number for this particular thing?Shawn Wang [00:14:31]: Yeah, it's retrieval. It's retrieval within machine learning. It's interesting because I think the more meta level I'm trying to operate at here is you have a benchmark. You're like, okay, I see the architectural thing I need to do in order to go fix that. But should you do it? Because sometimes that's an inductive bias, basically. It's what Jason Wei, who used to work at Google, would say. Exactly the kind of thing. Yeah, you're going to win. Short term. Longer term, I don't know if that's going to scale. You might have to undo that.Jeff Dean [00:15:01]: I mean, I like to sort of not focus on exactly what solution we're going to derive, but what capability would you want? And I think we're very convinced that, you know, long context is useful, but it's way too short today. Right? Like, I think what you would really want is, can I attend to the internet while I answer my question? Right? But that's not going to happen. I think that's going to be solved by purely scaling the existing solutions, which are quadratic. So a million tokens kind of pushes what you can do. You're not going to do that to a trillion tokens, let alone, you know, a billion tokens, let alone a trillion. But I think if you could give the illusion that you can attend to trillions of tokens, that would be amazing. You'd find all kinds of uses for that. You would have attend to the internet. You could attend to the pixels of YouTube and the sort of deeper representations that we can find. You could attend to the form for a single video, but across many videos, you know, on a personal Gemini level, you could attend to all of your personal state with your permission. So like your emails, your photos, your docs, your plane tickets you have. I think that would be really, really useful. And the question is, how do you get algorithmic improvements and system level improvements that get you to something where you actually can attend to trillions of tokens? Right. In a meaningful way. Yeah.Shawn Wang [00:16:26]: But by the way, I think I did some math and it's like, if you spoke all day, every day for eight hours a day, you only generate a maximum of like a hundred K tokens, which like very comfortably fits.Jeff Dean [00:16:38]: Right. But if you then say, okay, I want to be able to understand everything people are putting on videos.Shawn Wang [00:16:46]: Well, also, I think that the classic example is you start going beyond language into like proteins and whatever else is extremely information dense. Yeah. Yeah.Jeff Dean [00:16:55]: I mean, I think one of the things about Gemini's multimodal aspects is we've always wanted it to be multimodal from the start. And so, you know, that sometimes to people means text and images and video sort of human-like and audio, audio, human-like modalities. But I think it's also really useful to have Gemini know about non-human modalities. Yeah. Like LIDAR sensor data from. Yes. Say, Waymo vehicles or. Like robots or, you know, various kinds of health modalities, x-rays and MRIs and imaging and genomics information. And I think there's probably hundreds of modalities of data where you'd like the model to be able to at least be exposed to the fact that this is an interesting modality and has certain meaning in the world. Where even if you haven't trained on all the LIDAR data or MRI data, you could have, because maybe that's not, you know, it doesn't make sense in terms of trade-offs of. You know, what you include in your main pre-training data mix, at least including a little bit of it is actually quite useful. Yeah. Because it sort of tempts the model that this is a thing.Shawn Wang [00:18:04]: Yeah. Do you believe, I mean, since we're on this topic and something I just get to ask you all the questions I always wanted to ask, which is fantastic. Like, are there some king modalities, like modalities that supersede all the other modalities? So a simple example was Vision can, on a pixel level, encode text. And DeepSeq had this DeepSeq CR paper that did that. Vision. And Vision has also been shown to maybe incorporate audio because you can do audio spectrograms and that's, that's also like a Vision capable thing. Like, so, so maybe Vision is just the king modality and like. Yeah.Jeff Dean [00:18:36]: I mean, Vision and Motion are quite important things, right? Motion. Well, like video as opposed to static images, because I mean, there's a reason evolution has evolved eyes like 23 independent ways, because it's such a useful capability for sensing the world around you, which is really what we want these models to be. So I think the only thing that we can be able to do is interpret the things we're seeing or the things we're paying attention to and then help us in using that information to do things. Yeah.Shawn Wang [00:19:05]: I think motion, you know, I still want to shout out, I think Gemini, still the only native video understanding model that's out there. So I use it for YouTube all the time. Nice.Jeff Dean [00:19:15]: Yeah. Yeah. I mean, it's actually, I think people kind of are not necessarily aware of what the Gemini models can actually do. Yeah. Like I have an example I've used in one of my talks. It had like, it was like a YouTube highlight video of 18 memorable sports moments across the last 20 years or something. So it has like Michael Jordan hitting some jump shot at the end of the finals and, you know, some soccer goals and things like that. And you can literally just give it the video and say, can you please make me a table of what all these different events are? What when the date is when they happened? And a short description. And so you get like now an 18 row table of that information extracted from the video, which is, you know, not something most people think of as like a turn video into sequel like table.Alessio Fanelli [00:20:11]: Has there been any discussion inside of Google of like, you mentioned tending to the whole internet, right? Google, it's almost built because a human cannot tend to the whole internet and you need some sort of ranking to find what you need. Yep. That ranking is like much different for an LLM because you can expect a person to look at maybe the first five, six links in a Google search versus for an LLM. Should you expect to have 20 links that are highly relevant? Like how do you internally figure out, you know, how do we build the AI mode that is like maybe like much broader search and span versus like the more human one? Yeah.Jeff Dean [00:20:47]: I mean, I think even pre-language model based work, you know, our ranking systems would be built to start. I mean, I think even pre-language model based work, you know, our ranking systems would be built to start. With a giant number of web pages in our index, many of them are not relevant. So you identify a subset of them that are relevant with very lightweight kinds of methods. You know, you're down to like 30,000 documents or something. And then you gradually refine that to apply more and more sophisticated algorithms and more and more sophisticated sort of signals of various kinds in order to get down to ultimately what you show, which is, you know, the final 10 results or, you know, 10 results plus. Other kinds of information. And I think an LLM based system is not going to be that dissimilar, right? You're going to attend to trillions of tokens, but you're going to want to identify, you know, what are the 30,000 ish documents that are with the, you know, maybe 30 million interesting tokens. And then how do you go from that into what are the 117 documents I really should be paying attention to in order to carry out the tasks that the user has asked? And I think, you know, you can imagine systems where you have, you know, a lot of highly parallel processing to identify those initial 30,000 candidates, maybe with very lightweight kinds of models. Then you have some system that sort of helps you narrow down from 30,000 to the 117 with maybe a little bit more sophisticated model or set of models. And then maybe the final model is the thing that looks. So the 117 things that might be your most capable model. So I think it has to, it's going to be some system like that, that is really enables you to give the illusion of attending to trillions of tokens. Sort of the way Google search gives you, you know, not the illusion, but you are searching the internet, but you're finding, you know, a very small subset of things that are, that are relevant.Shawn Wang [00:22:47]: Yeah. I often tell a lot of people that are not steeped in like Google search history that, well, you know, like Bert was. Like he was like basically immediately inside of Google search and that improves results a lot, right? Like I don't, I don't have any numbers off the top of my head, but like, I'm sure you guys, that's obviously the most important numbers to Google. Yeah.Jeff Dean [00:23:08]: I mean, I think going to an LLM based representation of text and words and so on enables you to get out of the explicit hard notion of, of particular words having to be on the page, but really getting at the notion of this topic of this page or this page. Paragraph is highly relevant to this query. Yeah.Shawn Wang [00:23:28]: I don't think people understand how much LLMs have taken over all these very high traffic system, very high traffic. Yeah. Like it's Google, it's YouTube. YouTube has this like semantics ID thing where it's just like every token or every item in the vocab is a YouTube video or something that predicts the video using a code book, which is absurd to me for YouTube size.Jeff Dean [00:23:50]: And then most recently GROK also for, for XAI, which is like, yeah. I mean, I'll call out even before LLMs were used extensively in search, we put a lot of emphasis on softening the notion of what the user actually entered into the query.Shawn Wang [00:24:06]: So do you have like a history of like, what's the progression? Oh yeah.Jeff Dean [00:24:09]: I mean, I actually gave a talk in, uh, I guess, uh, web search and data mining conference in 2009, uh, where we never actually published any papers about the origins of Google search, uh, sort of, but we went through sort of four or five or six. generations, four or five or six generations of, uh, redesigning of the search and retrieval system, uh, from about 1999 through 2004 or five. And that talk is really about that evolution. And one of the things that really happened in 2001 was we were sort of working to scale the system in multiple dimensions. So one is we wanted to make our index bigger, so we could retrieve from a larger index, which always helps your quality in general. Uh, because if you don't have the page in your index, you're going to not do well. Um, and then we also needed to scale our capacity because we were, our traffic was growing quite extensively. Um, and so we had, you know, a sharded system where you have more and more shards as the index grows, you have like 30 shards. And then if you want to double the index size, you make 60 shards so that you can bound the latency by which you respond for any particular user query. Um, and then as traffic grows, you add, you add more and more replicas of each of those. And so we eventually did the math that realized that in a data center where we had say 60 shards and, um, you know, 20 copies of each shard, we now had 1200 machines, uh, with disks. And we did the math and we're like, Hey, one copy of that index would actually fit in memory across 1200 machines. So in 2001, we introduced, uh, we put our entire index in memory and what that enabled from a quality perspective was amazing. Um, and so we had more and more replicas of each of those. Before you had to be really careful about, you know, how many different terms you looked at for a query, because every one of them would involve a disk seek on every one of the 60 shards. And so you, as you make your index bigger, that becomes even more inefficient. But once you have the whole index in memory, it's totally fine to have 50 terms you throw into the query from the user's original three or four word query, because now you can add synonyms like restaurant and restaurants and cafe and, uh, you know, things like that. Uh, bistro and all these things. And you can suddenly start, uh, sort of really, uh, getting at the meaning of the word as opposed to the exact semantic form the user typed in. And that was, you know, 2001, very much pre LLM, but really it was about softening the, the strict definition of what the user typed in order to get at the meaning.Alessio Fanelli [00:26:47]: What are like principles that you use to like design the systems, especially when you have, I mean, in 2001, the internet is like. Doubling, tripling every year in size is not like, uh, you know, and I think today you kind of see that with LLMs too, where like every year the jumps in size and like capabilities are just so big. Are there just any, you know, principles that you use to like, think about this? Yeah.Jeff Dean [00:27:08]: I mean, I think, uh, you know, first, whenever you're designing a system, you want to understand what are the sort of design parameters that are going to be most important in designing that, you know? So, you know, how many queries per second do you need to handle? How big is the internet? How big is the index you need to handle? How much data do you need to keep for every document in the index? How are you going to look at it when you retrieve things? Um, what happens if traffic were to double or triple, you know, will that system work well? And I think a good design principle is you're going to want to design a system so that the most important characteristics could scale by like factors of five or 10, but probably not beyond that because often what happens is if you design a system for X. And something suddenly becomes a hundred X, that would enable a very different point in the design space that would not make sense at X. But all of a sudden at a hundred X makes total sense. So like going from a disk space index to a in memory index makes a lot of sense once you have enough traffic, because now you have enough replicas of the sort of state on disk that those machines now actually can hold, uh, you know, a full copy of the, uh, index and memory. Yeah. And that all of a sudden enabled. A completely different design that wouldn't have been practical before. Yeah. Um, so I'm, I'm a big fan of thinking through designs in your head, just kind of playing with the design space a little before you actually do a lot of writing of code. But, you know, as you said, in the early days of Google, we were growing the index, uh, quite extensively. We were growing the update rate of the index. So the update rate actually is the parameter that changed the most. Surprising. So it used to be once a month.Shawn Wang [00:28:55]: Yeah.Jeff Dean [00:28:56]: And then we went to a system that could update any particular page in like sub one minute. Okay.Shawn Wang [00:29:02]: Yeah. Because this is a competitive advantage, right?Jeff Dean [00:29:04]: Because all of a sudden news related queries, you know, if you're, if you've got last month's news index, it's not actually that useful for.Shawn Wang [00:29:11]: News is a special beast. Was there any, like you could have split it onto a separate system.Jeff Dean [00:29:15]: Well, we did. We launched a Google news product, but you also want news related queries that people type into the main index to also be sort of updated.Shawn Wang [00:29:23]: So, yeah, it's interesting. And then you have to like classify whether the page is, you have to decide which pages should be updated and what frequency. Oh yeah.Jeff Dean [00:29:30]: There's a whole like, uh, system behind the scenes that's trying to decide update rates and importance of the pages. So even if the update rate seems low, you might still want to recrawl important pages quite often because, uh, the likelihood they change might be low, but the value of having updated is high.Shawn Wang [00:29:50]: Yeah, yeah, yeah, yeah. Uh, well, you know, yeah. This, uh, you know, mention of latency and, and saving things to this reminds me of one of your classics, which I have to bring up, which is latency numbers. Every programmer should know, uh, was there a, was it just a, just a general story behind that? Did you like just write it down?Jeff Dean [00:30:06]: I mean, this has like sort of eight or 10 different kinds of metrics that are like, how long does a cache mistake? How long does branch mispredict take? How long does a reference domain memory take? How long does it take to send, you know, a packet from the U S to the Netherlands or something? Um,Shawn Wang [00:30:21]: why Netherlands, by the way, or is it, is that because of Chrome?Jeff Dean [00:30:25]: Uh, we had a data center in the Netherlands, um, so, I mean, I think this gets to the point of being able to do the back of the envelope calculations. So these are sort of the raw ingredients of those, and you can use them to say, okay, well, if I need to design a system to do image search and thumb nailing or something of the result page, you know, how, what I do that I could pre-compute the image thumbnails. I could like. Try to thumbnail them on the fly from the larger images. What would that do? How much dis bandwidth than I need? How many des seeks would I do? Um, and you can sort of actually do thought experiments in, you know, 30 seconds or a minute with the sort of, uh, basic, uh, basic numbers at your fingertips. Uh, and then as you sort of build software using higher level libraries, you kind of want to develop the same intuitions for how long does it take to, you know, look up something in this particular kind of.Shawn Wang [00:31:21]: I'll see you next time.Shawn Wang [00:31:51]: Which is a simple byte conversion. That's nothing interesting. I wonder if you have any, if you were to update your...Jeff Dean [00:31:58]: I mean, I think it's really good to think about calculations you're doing in a model, either for training or inference.Jeff Dean [00:32:09]: Often a good way to view that is how much state will you need to bring in from memory, either like on-chip SRAM or HBM from the accelerator. Attached memory or DRAM or over the network. And then how expensive is that data motion relative to the cost of, say, an actual multiply in the matrix multiply unit? And that cost is actually really, really low, right? Because it's order, depending on your precision, I think it's like sub one picodule.Shawn Wang [00:32:50]: Oh, okay. You measure it by energy. Yeah. Yeah.Jeff Dean [00:32:52]: Yeah. I mean, it's all going to be about energy and how do you make the most energy efficient system. And then moving data from the SRAM on the other side of the chip, not even off the off chip, but on the other side of the same chip can be, you know, a thousand picodules. Oh, yeah. And so all of a sudden, this is why your accelerators require batching. Because if you move, like, say, the parameter of a model from SRAM on the, on the chip into the multiplier unit, that's going to cost you a thousand picodules. So you better make use of that, that thing that you moved many, many times with. So that's where the batch dimension comes in. Because all of a sudden, you know, if you have a batch of 256 or something, that's not so bad. But if you have a batch of one, that's really not good.Shawn Wang [00:33:40]: Yeah. Yeah. Right.Jeff Dean [00:33:41]: Because then you paid a thousand picodules in order to do your one picodule multiply.Shawn Wang [00:33:46]: I have never heard an energy-based analysis of batching.Jeff Dean [00:33:50]: Yeah. I mean, that's why people batch. Yeah. Ideally, you'd like to use batch size one because the latency would be great.Shawn Wang [00:33:56]: The best latency.Jeff Dean [00:33:56]: But the energy cost and the compute cost inefficiency that you get is quite large. So, yeah.Shawn Wang [00:34:04]: Is there a similar trick like, like, like you did with, you know, putting everything in memory? Like, you know, I think obviously NVIDIA has caused a lot of waves with betting very hard on SRAM with Grok. I wonder if, like, that's something that you already saw with, with the TPUs, right? Like that, that you had to. Uh, to serve at your scale, uh, you probably sort of saw that coming. Like what, what, what hardware, uh, innovations or insights were formed because of what you're seeing there?Jeff Dean [00:34:33]: Yeah. I mean, I think, you know, TPUs have this nice, uh, sort of regular structure of 2D or 3D meshes with a bunch of chips connected. Yeah. And each one of those has HBM attached. Um, I think for serving some kinds of models, uh, you know, you, you pay a lot higher cost. Uh, and time latency, um, bringing things in from HBM than you do bringing them in from, uh, SRAM on the chip. So if you have a small enough model, you can actually do model parallelism, spread it out over lots of chips and you actually get quite good throughput improvements and latency improvements from doing that. And so you're now sort of striping your smallish scale model over say 16 or 64 chips. Uh, but as if you do that and it all fits in. In SRAM, uh, that can be a big win. So yeah, that's not a surprise, but it is a good technique.Alessio Fanelli [00:35:27]: Yeah. What about the TPU design? Like how much do you decide where the improvements have to go? So like, this is like a good example of like, is there a way to bring the thousand picojoules down to 50? Like, is it worth designing a new chip to do that? The extreme is like when people say, oh, you should burn the model on the ASIC and that's kind of like the most extreme thing. How much of it? Is it worth doing an hardware when things change so quickly? Like what was the internal discussion? Yeah.Jeff Dean [00:35:57]: I mean, we, we have a lot of interaction between say the TPU chip design architecture team and the sort of higher level modeling, uh, experts, because you really want to take advantage of being able to co-design what should future TPUs look like based on where we think the sort of ML research puck is going, uh, in some sense, because, uh, you know, as a hardware designer for ML and in particular, you're trying to design a chip starting today and that design might take two years before it even lands in a data center. And then it has to sort of be a reasonable lifetime of the chip to take you three, four or five years. So you're trying to predict two to six years out where, what ML computations will people want to run two to six years out in a very fast changing field. And so having people with interest. Interesting ML research ideas of things we think will start to work in that timeframe or will be more important in that timeframe, uh, really enables us to then get, you know, interesting hardware features put into, you know, TPU N plus two, where TPU N is what we have today.Shawn Wang [00:37:10]: Oh, the cycle time is plus two.Jeff Dean [00:37:12]: Roughly. Wow. Because, uh, I mean, sometimes you can squeeze some changes into N plus one, but, you know, bigger changes are going to require the chip. Yeah. Design be earlier in its lifetime design process. Um, so whenever we can do that, it's generally good. And sometimes you can put in speculative features that maybe won't cost you much chip area, but if it works out, it would make something, you know, 10 times as fast. And if it doesn't work out, well, you burned a little bit of tiny amount of your chip area on that thing, but it's not that big a deal. Uh, sometimes it's a very big change and we want to be pretty sure this is going to work out. So we'll do like lots of carefulness. Uh, ML experimentation to show us, uh, this is actually the, the way we want to go. Yeah.Alessio Fanelli [00:37:58]: Is there a reverse of like, we already committed to this chip design so we can not take the model architecture that way because it doesn't quite fit?Jeff Dean [00:38:06]: Yeah. I mean, you, you definitely have things where you're going to adapt what the model architecture looks like so that they're efficient on the chips that you're going to have for both training and inference of that, of that, uh, generation of model. So I think it kind of goes both ways. Um, you know, sometimes you can take advantage of, you know, lower precision things that are coming in a future generation. So you can, might train it at that lower precision, even if the current generation doesn't quite do that. Mm.Shawn Wang [00:38:40]: Yeah. How low can we go in precision?Jeff Dean [00:38:43]: Because people are saying like ternary is like, uh, yeah, I mean, I'm a big fan of very low precision because I think that gets, that saves you a tremendous amount of time. Right. Because it's picojoules per bit that you're transferring and reducing the number of bits is a really good way to, to reduce that. Um, you know, I think people have gotten a lot of luck, uh, mileage out of having very low bit precision things, but then having scaling factors that apply to a whole bunch of, uh, those, those weights. Scaling. How does it, how does it, okay.Shawn Wang [00:39:15]: Interesting. You, so low, low precision, but scaled up weights. Yeah. Huh. Yeah. Never considered that. Yeah. Interesting. Uh, w w while we're on this topic, you know, I think there's a lot of, um, uh, this, the concept of precision at all is weird when we're sampling, you know, uh, we just, at the end of this, we're going to have all these like chips that I'll do like very good math. And then we're just going to throw a random number generator at the start. So, I mean, there's a movement towards, uh, energy based, uh, models and processors. I'm just curious if you've, obviously you've thought about it, but like, what's your commentary?Jeff Dean [00:39:50]: Yeah. I mean, I think. There's a bunch of interesting trends though. Energy based models is one, you know, diffusion based models, which don't sort of sequentially decode tokens is another, um, you know, speculative decoding is a way that you can get sort of an equivalent, very small.Shawn Wang [00:40:06]: Draft.Jeff Dean [00:40:07]: Batch factor, uh, for like you predict eight tokens out and that enables you to sort of increase the effective batch size of what you're doing by a factor of eight, even, and then you maybe accept five or six of those tokens. So you get. A five, a five X improvement in the amortization of moving weights, uh, into the multipliers to do the prediction for the, the tokens. So these are all really good techniques and I think it's really good to look at them from the lens of, uh, energy, real energy, not energy based models, um, and, and also latency and throughput, right? If you look at things from that lens, that sort of guides you to. Two solutions that are gonna be, uh, you know, better from, uh, you know, being able to serve larger models or, you know, equivalent size models more cheaply and with lower latency.Shawn Wang [00:41:03]: Yeah. Well, I think, I think I, um, it's appealing intellectually, uh, haven't seen it like really hit the mainstream, but, um, I do think that, uh, there's some poetry in the sense that, uh, you know, we don't have to do, uh, a lot of shenanigans if like we fundamentally. Design it into the hardware. Yeah, yeah.Jeff Dean [00:41:23]: I mean, I think there's still a, there's also sort of the more exotic things like analog based, uh, uh, computing substrates as opposed to digital ones. Uh, I'm, you know, I think those are super interesting cause they can be potentially low power. Uh, but I think you often end up wanting to interface that with digital systems and you end up losing a lot of the power advantages in the digital to analog and analog to digital conversions. You end up doing, uh, at the sort of boundaries. And periphery of that system. Um, I still think there's a tremendous distance we can go from where we are today in terms of energy efficiency with sort of, uh, much better and specialized hardware for the models we care about.Shawn Wang [00:42:05]: Yeah.Alessio Fanelli [00:42:06]: Um, any other interesting research ideas that you've seen, or like maybe things that you cannot pursue a Google that you would be interested in seeing researchers take a step at, I guess you have a lot of researchers. Yeah, I guess you have enough, but our, our research.Jeff Dean [00:42:21]: Our research portfolio is pretty broad. I would say, um, I mean, I think, uh, in terms of research directions, there's a whole bunch of, uh, you know, open problems and how do you make these models reliable and able to do much longer, kind of, uh, more complex tasks that have lots of subtasks. How do you orchestrate, you know, maybe one model that's using other models as tools in order to sort of build, uh, things that can accomplish, uh, you know, much more. Yeah. Significant pieces of work, uh, collectively, then you would ask a single model to do. Um, so that's super interesting. How do you get more verifiable, uh, you know, how do you get RL to work for non-verifiable domains? I think it's a pretty interesting open problem because I think that would broaden out the capabilities of the models, the improvements that you're seeing in both math and coding. Uh, if we could apply those to other less verifiable domains, because we've come up with RL techniques that actually enable us to do that. Uh, effectively, that would, that would really make the models improve quite a lot. I think.Alessio Fanelli [00:43:26]: I'm curious, like when we had Noam Brown on the podcast, he said, um, they already proved you can do it with deep research. Um, you kind of have it with AI mode in a way it's not verifiable. I'm curious if there's any thread that you think is interesting there. Like what is it? Both are like information retrieval of JSON. So I wonder if it's like the retrieval is like the verifiable part. That you can score or what are like, yeah, yeah. How, how would you model that, that problem?Jeff Dean [00:43:55]: Yeah. I mean, I think there are ways of having other models that can evaluate the results of what a first model did, maybe even retrieving. Can you have another model that says, is this things, are these things you retrieved relevant? Or can you rate these 2000 things you retrieved to assess which ones are the 50 most relevant or something? Um, I think those kinds of techniques are actually quite effective. Sometimes I can even be the same model, just prompted differently to be a, you know, a critic as opposed to a, uh, actual retrieval system. Yeah.Shawn Wang [00:44:28]: Um, I do think like there, there is that, that weird cliff where like, it feels like we've done the easy stuff and then now it's, but it always feels like that every year. It's like, oh, like we know, we know, and the next part is super hard and nobody's figured it out. And, uh, exactly with this RLVR thing where like everyone's talking about, well, okay, how do we. the next stage of the non-verifiable stuff. And everyone's like, I don't know, you know, Ellen judge.Jeff Dean [00:44:56]: I mean, I feel like the nice thing about this field is there's lots and lots of smart people thinking about creative solutions to some of the problems that we all see. Uh, because I think everyone sort of sees that the models, you know, are great at some things and they fall down around the edges of those things and, and are not as capable as we'd like in those areas. And then coming up with good techniques and trying those. And seeing which ones actually make a difference is sort of what the whole research aspect of this field is, is pushing forward. And I think that's why it's super interesting. You know, if you think about two years ago, we were struggling with GSM, eight K problems, right? Like, you know, Fred has two rabbits. He gets three more rabbits. How many rabbits does he have? That's a pretty far cry from the kinds of mathematics that the models can, and now you're doing IMO and Erdos problems in pure language. Yeah. Yeah. Pure language. So that is a really, really amazing jump in capabilities in, you know, in a year and a half or something. And I think, um, for other areas, it'd be great if we could make that kind of leap. Uh, and you know, we don't exactly see how to do it for some, some areas, but we do see it for some other areas and we're going to work hard on making that better. Yeah.Shawn Wang [00:46:13]: Yeah.Alessio Fanelli [00:46:14]: Like YouTube thumbnail generation. That would be very helpful. We need that. That would be AGI. We need that.Shawn Wang [00:46:20]: That would be. As far as content creators go.Jeff Dean [00:46:22]: I guess I'm not a YouTube creator, so I don't care that much about that problem, but I guess, uh, many people do.Shawn Wang [00:46:27]: It does. Yeah. It doesn't, it doesn't matter. People do judge books by their covers as it turns out. Um, uh, just to draw a bit on the IMO goal. Um, I'm still not over the fact that a year ago we had alpha proof and alpha geometry and all those things. And then this year we were like, screw that we'll just chuck it into Gemini. Yeah. What's your reflection? Like, I think this, this question about. Like the merger of like symbolic systems and like, and, and LMS, uh, was a very much core belief. And then somewhere along the line, people would just said, Nope, we'll just all do it in the LLM.Jeff Dean [00:47:02]: Yeah. I mean, I think it makes a lot of sense to me because, you know, humans manipulate symbols, but we probably don't have like a symbolic representation in our heads. Right. We have some distributed representation that is neural net, like in some way of lots of different neurons. And activation patterns firing when we see certain things and that enables us to reason and plan and, you know, do chains of thought and, you know, roll them back now that, that approach for solving the problem doesn't seem like it's going to work. I'm going to try this one. And, you know, in a lot of ways we're emulating what we intuitively think, uh, is happening inside real brains in neural net based models. So it never made sense to me to have like completely separate. Uh, discrete, uh, symbolic things, and then a completely different way of, of, uh, you know, thinking about those things.Shawn Wang [00:47:59]: Interesting. Yeah. Uh, I mean, it's maybe seems obvious to you, but it wasn't obvious to me a year ago. Yeah.Jeff Dean [00:48:06]: I mean, I do think like that IMO with, you know, translating to lean and using lean and then the next year and also a specialized geometry model. And then this year switching to a single unified model. That is roughly the production model with a little bit more inference budget, uh, is actually, you know, quite good because it shows you that the capabilities of that general model have improved dramatically and, and now you don't need the specialized model. This is actually sort of very similar to the 2013 to 16 era of machine learning, right? Like it used to be, people would train separate models for lots of different, each different problem, right? I have, I want to recognize street signs and something. So I train a street sign. Recognition recognition model, or I want to, you know, decode speech recognition. I have a speech model, right? I think now the era of unified models that do everything is really upon us. And the question is how well do those models generalize to new things they've never been asked to do and they're getting better and better.Shawn Wang [00:49:10]: And you don't need domain experts. Like one of my, uh, so I interviewed ETA who was on, who was on that team. Uh, and he was like, yeah, I, I don't know how they work. I don't know where the IMO competition was held. I don't know the rules of it. I just trained the models, the training models. Yeah. Yeah. And it's kind of interesting that like people with these, this like universal skill set of just like machine learning, you just give them data and give them enough compute and they can kind of tackle any task, which is the bitter lesson, I guess. I don't know. Yeah.Jeff Dean [00:49:39]: I mean, I think, uh, general models, uh, will win out over specialized ones in most cases.Shawn Wang [00:49:45]: Uh, so I want to push there a bit. I think there's one hole here, which is like, uh. There's this concept of like, uh, maybe capacity of a model, like abstractly a model can only contain the number of bits that it has. And, uh, and so it, you know, God knows like Gemini pro is like one to 10 trillion parameters. We don't know, but, uh, the Gemma models, for example, right? Like a lot of people want like the open source local models that are like that, that, that, and, and, uh, they have some knowledge, which is not necessary, right? Like they can't know everything like, like you have the. The luxury of you have the big model and big model should be able to capable of everything. But like when, when you're distilling and you're going down to the small models, you know, you're actually memorizing things that are not useful. Yeah. And so like, how do we, I guess, do we want to extract that? Can we, can we divorce knowledge from reasoning, you know?Jeff Dean [00:50:38]: Yeah. I mean, I think you do want the model to be most effective at reasoning if it can retrieve things, right? Because having the model devote precious parameter space. To remembering obscure facts that could be looked up is actually not the best use of that parameter space, right? Like you might prefer something that is more generally useful in more settings than this obscure fact that it has. Um, so I think that's always attention at the same time. You also don't want your model to be kind of completely detached from, you know, knowing stuff about the world, right? Like it's probably useful to know how long the golden gate be. Bridges just as a general sense of like how long are bridges, right? And, uh, it should have that kind of knowledge. It maybe doesn't need to know how long some teeny little bridge in some other more obscure part of the world is, but, uh, it does help it to have a fair bit of world knowledge and the bigger your model is, the more you can have. Uh, but I do think combining retrieval with sort of reasoning and making the model really good at doing multiple stages of retrieval. Yeah.Shawn Wang [00:51:49]: And reasoning through the intermediate retrieval results is going to be a, a pretty effective way of making the model seem much more capable, because if you think about, say, a personal Gemini, yeah, right?Jeff Dean [00:52:01]: Like we're not going to train Gemini on my email. Probably we'd rather have a single model that, uh, we can then use and use being able to retrieve from my email as a tool and have the model reason about it and retrieve from my photos or whatever, uh, and then make use of that and have multiple. Um, you know, uh, stages of interaction. that makes sense.Alessio Fanelli [00:52:24]: Do you think the vertical models are like, uh, interesting pursuit? Like when people are like, oh, we're building the best healthcare LLM, we're building the best law LLM, are those kind of like short-term stopgaps or?Jeff Dean [00:52:37]: No, I mean, I think, I think vertical models are interesting. Like you want them to start from a pretty good base model, but then you can sort of, uh, sort of viewing them, view them as enriching the data. Data distribution for that particular vertical domain for healthcare, say, um, we're probably not going to train or for say robotics. We're probably not going to train Gemini on all possible robotics data. We, you could train it on because we want it to have a balanced set of capabilities. Um, so we'll expose it to some robotics data, but if you're trying to build a really, really good robotics model, you're going to want to start with that and then train it on more robotics data. And then maybe that would. It's multilingual translation capability, but improve its robotics capabilities. And we're always making these kind of, uh, you know, trade-offs in the data mix that we train the base Gemini models on. You know, we'd love to include data from 200 more languages and as much data as we have for those languages, but that's going to displace some other capabilities of the model. It won't be as good at, um, you know, Pearl programming, you know, it'll still be good at Python programming. Cause we'll include it. Enough. Of that, but there's other long tail computer languages or coding capabilities that it may suffer on or multi, uh, multimodal reasoning capabilities may suffer. Cause we didn't get to expose it to as much data there, but it's really good at multilingual things. So I, I think some combination of specialized models, maybe more modular models. So it'd be nice to have the capability to have those 200 languages, plus this awesome robotics model, plus this awesome healthcare, uh, module that all can be knitted together to work in concert and called upon in different circumstances. Right? Like if I have a health related thing, then it should enable using this health module in conjunction with the main base model to be even better at those kinds of things. Yeah.Shawn Wang [00:54:36]: Installable knowledge. Yeah.Jeff Dean [00:54:37]: Right.Shawn Wang [00:54:38]: Just download as a, as a package.Jeff Dean [00:54:39]: And some of that installable stuff can come from retrieval, but some of it probably should come from preloaded training on, you know, uh, a hundred billion tokens or a trillion tokens of health data. Yeah.Shawn Wang [00:54:51]: And for listeners, I think, uh, I will highlight the Gemma three end paper where they, there was a little bit of that, I think. Yeah.Alessio Fanelli [00:54:56]: Yeah. I guess the question is like, how many billions of tokens do you need to outpace the frontier model improvements? You know, it's like, if I have to make this model better healthcare and the main. Gemini model is still improving. Do I need 50 billion tokens? Can I do it with a hundred, if I need a trillion healthcare tokens, it's like, they're probably not out there that you don't have, you know, I think that's really like the.Jeff Dean [00:55:21]: Well, I mean, I think healthcare is a particularly challenging domain, so there's a lot of healthcare data that, you know, we don't have access to appropriately, but there's a lot of, you know, uh, healthcare organizations that want to train models on their own data. That is not public healthcare data, uh, not public health. But public healthcare data. Um, so I think there are opportunities there to say, partner with a large healthcare organization and train models for their use that are going to be, you know, more bespoke, but probably, uh, might be better than a general model trained on say, public data. Yeah.Shawn Wang [00:55:58]: Yeah. I, I believe, uh, by the way, also this is like somewhat related to the language conversation. Uh, I think one of your, your favorite examples was you can put a low resource language in the context and it just learns. Yeah.Jeff Dean [00:56:09]: Oh, yeah, I think the example we used was Calamon, which is truly low resource because it's only spoken by, I think 120 people in the world and there's no written text.Shawn Wang [00:56:20]: So, yeah. So you can just do it that way. Just put it in the context. Yeah. Yeah. But I think your whole data set in the context, right.Jeff Dean [00:56:27]: If you, if you take a language like, uh, you know, Somali or something, there is a fair bit of Somali text in the world that, uh, or Ethiopian Amharic or something, um, you know, we probably. Yeah. Are not putting all the data from those languages into the Gemini based training. We put some of it, but if you put more of it, you'll improve the capabilities of those models.Shawn Wang [00:56:49]: Yeah.Jeff Dean [00:56:49]:

La segunda jornada de pruebas en Baréin ha vuelto a poner de manifiesto que esta pretemporada está definida por la adaptación a la nueva normativa híbrida y por una fiabilidad todavía frágil en varios equipos. El Podcast Técnica Fórmula 1 nos hace un análisis profundo de lo que está ocurriendo en esta primera semana de pretemporada. Unos bien y otros mal Aunque el día arrancó con sobresaltos, también dejó las primeras señales claras del rendimiento de algunos motores, especialmente el Ferrari, y el esperado regreso de Red Bull a pista tras una mañana prácticamente en blanco. La sesión matinal comenzó con una bandera roja temprana protagonizada por Sergio Pérez, cuyo Cadillac se detuvo alrededor de las 8:15. El incidente, no obstante, quedó en un susto: antes de las nueve ya estaba completando vueltas. Quien sí tardó en aparecer fue Red Bull, que permaneció en el box debido a un problema rutinario detectado durante el montaje nocturno del RB20. No rodaron hasta el último minuto de la mañana, donde Isack Hadjar se limitó a hacer una vuelta de instalación. Los problemas se extendieron también a Aston Martin, nuevamente protagonista negativo del día. Fernando Alonso rodó poco por la tarde y el equipo reconoció abiertamente la magnitud de su déficit. Después de acumular cerca de 400 kilómetros menos que la mayoría de sus rivales (sumando Barcelona y Baréin), Lance Stroll fue contundente: “Tenemos que seguir empujando. Son problemas de motor, de equilibrio, de grip... es una combinación de cosas”, admitió. El canadiense incluso llegó a situar su desventaja en torno a cuatro o cuatro segundos y medio por vuelta, aunque subrayó que es imposible conocer la carga de combustible del resto. La sesión vespertina también dejó otras incidencias: una pieza desprendida del Audi R26 de Gabriel Bortoleto provocó una bandera roja a las 14:35, seguida poco después por otra neutralización debido a problemas técnicos en el Alpine de Pierre Gasly. La polémica del motor Mercedes. Además, el día ha estado marcado por nuevas informaciones sobre el controvertido motor Mercedes de 2026: se empieza a perfilar un acuerdo temporal entre FIA, FOM y los fabricantes para limitar el famoso “truco” sin prohibirlo por completo, a la espera de una resolución definitiva en 2027. En pista, las escuderías han continuado centradas en el aprendizaje energético. Durante las primeras horas se vieron tandas cortas de 5 a 6 vueltas, destinadas a calibrar la recuperación, los mapas motor y el comportamiento del lift and coast. El caso más llamativo ha sido el de Charles Leclerc, que alcanzó los 324 km/h para luego levantar claramente antes del final de recta. El motor Ferrari demostró una velocidad punta muy competitiva y, sobre todo, una consistencia que empieza a llamar la atención de todo el paddock. Mucho más kilometraje. Con el paso del día los equipos han aumentado el kilometraje con stints de entre 11 y 17 vueltas. Destacan las tandas de Norris, Hülkenberg, Gasly, Lawson y Albon, mientras Alonso completaba más vueltas que en todo el día anterior, aunque siempre en programas alejados del rendimiento puro y centrados en la gestión de energía. Las dificultades de pilotaje volvieron a ser protagonistas: numerosos pilotos sufrieron salidas de pista debido a comportamientos impredecibles al soltar la potencia, un síntoma de lo exigente que será el nuevo reglamento. Carlos Sainz lo resumió con claridad: “Los coches cambian el punto de frenada de una vuelta a otra”. Alguna situación curiosa. La coordinación entre equipos también ha sido visible hoy. Norris y Albon, ambos con motor Mercedes, estuvieron rodando juntos durante varias vueltas para probar el modo adelantamiento, siguiéndose de cerca en la recta principal para evaluar diferencias de despliegue eléctrico. El stint más largo del día volvió a ser obra de Norris, con 15 vueltas consecutivas. Por otra parte, Red Bull ha cerrado la jornada con mejor sabor de boca: por la tarde, Hadjar acumuló rodaje real y Verstappen tomará el relevo mañana. En Aston Martin, pese a las pocas vueltas de Alonso, hubo un detalle alentador: el asturiano salió en su último stint con velocidad y luces encendidas, y el coche más cerrado, señal de que el equipo empieza a desbloquear parte de la potencia del AMR26, aparentemente sin riesgo de rotura, sino por ajustes de software y electrónica. Finalmente, en el plano aerodinámico, el día no nos ha dejado novedades significativas, aunque sí se vieron parrillas de sensores en Racing Bulls, McLaren y Ferrari, además de parafina en el Audi. Lo más interesante: los difusores y otras áreas ocultas reveladas durante las pausas, que serán analizadas en detalle en el artículo técnico del viernes. La tercera jornada contará con rotaciones habituales: Norris y Piastri en McLaren, Leclerc por la mañana con Ferrari y Hamilton por la tarde, Russell y Antonelli dividiéndose el Mercedes, y Verstappen regresando en Red Bull antes de ceder el coche nuevamente a Hadjar. Con dos días completados y sólo uno por delante, las escuderías afrontan ya la recta final de estos decisivos tests de 2026. Escucha el episodio completo en la app de iVoox, o descubre todo el catálogo de iVoox Originals

En este programa te hablo de la marca francesa Valone, que se presenta con el modelo de trail Rave. Una zapatilla (según la marca, no la he probado) de media-larga distancia.Esta zapatilla tiene un upper de kevlar y un paquete de suela-mediasuela Vibram. Ofrece un drop 7 y una suela Megragrip con Traction Lug y tacos de 4'5mm.Unos de los grandes activos, es la "reparabilidad". Y es que la marca, en su propia web ya ofrece la posibilidad de tramitar el intercambio de suela-mediasuela, de forma simple y cómoda, con un precio de 100€.Eso sí, su único modelo Rave (disponible en 3 colores y hasta el número 46'5) tiene un precio de 219€.Me parece un producto curioso e interesante, y de él te hablo en este programa.Esta es su web, por si quieres darle un vistazo:https://www.valonerun.com/Contacto:juan@ellaboratoriodejuan.com

Unos consideran que ahorcamiento de EU que deja sin luz, sin gasolina, sin agua y sin alimentos a isleños provocará una migración masiva

Parece una bola esponjosa inocente en el agua. ¡Pero cuidado! Lo que estás mirando es un erizo de mar, ¡y es algo de lo que quieres estar muy, muy lejos! Esa "pelusa" son en realidad púas puntiagudas y afiladas llenas de veneno. Cuando te encuentras con estas cosas, a menudo es demasiado tarde: ¡ya lo pisaste y tienes segundos para reaccionar! En caso de que eso suceda, ¡sigue mirando para ver qué debes hacer! Los erizos no son las únicas criaturas venenosas que se esconden en tu playa local. De nuevo, probablemente ni siquiera verás el pez piedra hasta que su veneno ya esté en ti por un paso mal dado. Si su púa se mete en ti, ¡necesitarás ayuda rápidamente! ¡Cuidado, hay muchas más criaturas marinas aterradoras que debes conocer la próxima vez que vayas a la playa! ? Learn more about your ad choices. Visit megaphone.fm/adchoices

'La Fiera' es una película que narra la historia de estos cinco amigos unidos por el salto base y marcados por una sucesión de tragedias que transformaron su vínculo y su forma de entender la vida.

Durante las últimas semanas, el debate sobre los carbohidratos en el rendimiento de resistencia ha vuelto a estallar. Unos defienden que son imprescindibles para rendir; otros, que son innecesarios e incluso perjudiciales. El problema es que ambos bandos están mirando solo una parte del sistema. En este artículo no voy a tomar partido por un lado u otro, sino ordenar el caos. Porque cuando entiendes el contexto completo, el debate deja de ser confuso y empieza a tener sentido. *** Lista para avisos sobre el libro: https://solaarjona.com/lista/ *** Enlaces del vídeo: 208. Salud bucodental y rendimiento, con Eider Unamuno https://www.youtube.com/watch?v=P8dRo9z0CyU Ciclismo Evolutivo - 187. Microbiota y sistema inmunitario en el rendimiento, con Sari Arponen. https://www.youtube.com/watch?v=PjmSsfwqkUU Ciclismo Evolutivo - 189. Una visión integral del rendimiento deportivo, con Jesús Álvarez-Herms. https://www.youtube.com/watch?v=S1oPq5zfWoY 238. Pros y contras de los hidratos de carbono en el rendimiento. https://youtu.be/cSvlNbBnG3c?si=2joktrRz8WKtynaE *** https://solaarjona.com/lista/

La directora Isabel Coixet nos recibe en su estudio de Barcelona. Unos bajos con un pequeño jardín en el barrio de Gràcia de Barcelona. Un espacio presidido por la directora Agnès Varda, en una figura tamaño natural de cartón troquelado y fotos que la misma Coixet le hizo. Confiesa que mantiene conversaciones con ella, hablamos de la belleza de la vida y de esas pequeñas cosas que nos salva. Eso es parte del tema de su última película "los tres adioses" que se acaba de estrenar en cines.

El calendario electoral de este 2026 comenzó con la reciente elección de Laura Fernández en primera vuelta en Costa Rica. Unos comicios en los que más allá del contundente resultado a favor del partido oficial, destaca la solidez, transparencia y profesionalismo institucional del proceso costarricense tal como los confirma la observación electoral internacional. Y es precisamente en el papel que desempeña este acompañamiento técnico externo, es en donde queremos fijar nuestro análisis en este episodio de Mirada Semanal. Destacando su importancia para el mejoramiento de la integridad electoral de una región marcada por la polarización, el avance del extremismo y la sostenida desafección electoral de importantes segmentos de la ciudadanía.Y es que además de Costa Rica, este año será decisivo para países con grandes niveles de conflictividad social, violencia política y permanentes tensiones institucionales como Colombia, Perú, Brasil o Haití, sin descartarse la posibilidad de una convocatoria electoral sobrevenida en Venezuela. Cabe destacar el caso de Brasil, la mayor economía del continente y una de las más influyentes diplomáticamente, celebrará comicios generales en octubre. Un proceso comicial que pondrá a prueba si la gestión de Lula Da Silva ha podido realmente recomponer el tejido social y superar el sectarismo, la corrupción y la crispación política del pasado recientes. También destaca el caso de Colombia, una nación con un historial reciente de conflicto interno, asesinatos políticos y debates profundos sobre paz y justicia, en el que sus ciudadanos votarán en mayo para elegir un nuevo presidente en medio de una feroz contienda política. Más al sur, Perú nación que acudirá a elecciones en abril en un contexto de alta volatilidad institucional, fragmentación partidista tras años de crisis institucional que han generado descontento generalizado y sostenido en el país andino. Sin embargo, el caso más crónico será el de Haití, el proceso electoral de agosto buscará reestablecer una mínima normalidad tras años de violencia entre pandillas, desplazados, pobreza y una crisis humanitaria continuada.En virtud de la compleja agenda política que debe afrontarse en buena parte de los países de la región, la observación electoral resulta crucial. Especialmente para promover la transparencia, documentar irregularidades y ofrecer recomendaciones técnicas de lo que ocurre en cada etapa del proceso, desde la campaña hasta la comunicación de resultados. En un contexto regional donde el descrédito de la política ha alimentado teorías de fraude y desconfianza generalizada en el proceso electoral, la labor de observadores puede marcar la diferencia entre un resultado aceptado socialmente o uno que avive más las tensiones. Para analizar este tema nos acompañó la Dra. Ana Claudia Santano. Doctora y Máster en Ciencias Jurídicas y Políticas de la Universidad de Salamanca, España. Profesora de Derecho Constitucional, Electoral y Derechos Humanos en diversas instituciones de Brasil y América Latina. Coordinadora general de Transparencia Electoral Brasil. Una voz calificada desde la experiencia internacional quien nos ayudará a entender la observación electoral en tiempos de deterioro democrático.Analistas:Manuel Alcántara SáezMaría Puerta RieraInvitada:Ana Cláudia SantanoEdición y Conducción:Xavier Rodríguez Franco

Meditación del Evangelio según San Marcos 6, 14-29 por el biblista P. Norberto Padilla, misionero claretiano.Viernes 6/feb/2026, Es Juan, a quien yo decapité, que ha resucitado.Canción: Hazme ver (2017), de José Ibáñez----------Lectura del santo evangelio según san Marcos 6, 14-29En aquel tiempo, como la fama de Jesús se había extendido, el rey Herodes oyó hablar de él. Unos decían: «Juan Bautista ha resucitado, y por eso los poderes actúan en él.» Otros decían: «Es Elías.» Otros: «Es un profeta como los antiguos.» Herodes, al oírlo, decía: «Es Juan, a quien yo decapité, que ha resucitado.» Es que Herodes había mandado prender a Juan y lo había metido en la cárcel, encadenado. El motivo era que Herodes se había casado con Herodías, mujer de su hermano Filipo, y Juan le decía que no le era lícito tener la mujer de su hermano. Herodías aborrecía a Juan y quería quitarlo de en medio; no acababa de conseguirlo, porque Herodes respetaba a Juan, sabiendo que era un hombre honrado y santo, y lo defendía. Cuando lo escuchaba, quedaba desconcertado, y lo escuchaba con gusto. La ocasión llegó cuando Herodes, por su cumpleaño, dio un banquete a sus magnates, a sus oficiales y a la gente principal de Galilea. La hija de Herodías entró y danzó, gustando mucho a Herodes y a los convidados. El rey le dijo a la joven: «Pídeme lo que quieras, que te lo doy.» Y le juró: «Te daré lo que me pidas, aunque sea la mitad de mi reino.» Ella salió a preguntarle a su madre: «¿Qué le pido?» La madre le contestó: «La cabeza de Juan el Bautista.» Entró ella en seguida, a toda prisa, se acercó al rey y le pidió: «Quiero que ahora mismo me des en una bandeja la cabeza de Juan, el Bautista.» El rey se puso muy triste; pero, por el juramento y los convidados, no quiso desairarla. En seguida mandó a un verdugo que trajese la cabeza de Juan. Fue, lo decapitó en la cárcel, trajo la cabeza en una bandeja y se la entregó a la joven; la joven se la entregó a su madre. Al enterarse sus discípulos, fueron a recoger el cadáver y lo enterraron.Palabra del Señor... Gloria a ti, Señor Jesús#SoyClaretiano #Evangelio #MisionerosClaretianos #CMFAntillasIntro: Lámpara Es Tu Palabra, de Ain Karem

“Pues el Señor mismo descenderá del cielo con un grito de mando, con voz de arcángel y con el llamado de trompeta de Dios. Primero, los creyentes que hayan muerto se levantarán de sus tumbas. Luego, junto con ellos, nosotros, los que aún sigamos vivos sobre la tierra, seremos arrebatados en las nubes para encontrarnos con el Señor en el aire. Entonces estaremos con el Señor para siempre. Así que anímense unos a otros con estas palabras.”1 Tesalonicenses‬ ‭4‬:‭16‬-‭18‬ ‭NTV‬‬

Muchos más recursos para tu vida de fe (Santo Rosario, Oración, etc.) en nuestra web https://sercreyente.com________________Viernes, 6 de febrero de 2026 (4ª Semana del Tiempo Ordinario)Evangelio del día y reflexión... ¡Deja que la Palabra del Señor transforme tu vida! Texto íntegro del Evangelio y de la Reflexión en https://sercreyente.com/no-desprecian-a-un-profeta-mas-que-en-su-tierra/[Marcos 6, 14-29] En aquel tiempo, como la fama de Jesús se había extendido, el rey Herodes oyó hablar de él. Unos decían: «Juan el Bautista ha resucitado de entre los muertos y por eso las fuerzas milagrosas actúan en él». Otros decían: «Es Elías». Otros: «Es un profeta como los antiguos». Herodes, al oírlo, decía: «Es Juan, a quien yo decapité, que ha resucitado». Es que Herodes había mandado prender a Juan y lo había metido en la cárcel encadenado. El motivo era que Herodes se había casado con Herodías, mujer de su hermano Filipo, y Juan le decía que no le era lícito tener a la mujer de su hermano. Herodías aborrecía a Juan y quería matarlo, pero no podía, porque Herodes respetaba a Juan, sabiendo que era un hombre justo y santo, y lo defendía. Al escucharlo quedaba muy perplejo, aunque lo oía con gusto. La ocasión llegó cuando Herodes, por su cumpleaños, dio un banquete a sus magnates, a sus oficiales y a la gente principal de Galilea. La hija de Herodías entró y danzó, gustando mucho a Herodes y a los convidados. El rey le dijo a la joven: «Pídeme lo que quieras, que te lo daré». Y le juró: «Te daré lo que me pidas, aunque sea la mitad de mi reino». Ella salió a preguntarle a su madre: «¿Qué le pido?». La madre le contestó: «La cabeza de Juan el Bautista». Entró ella enseguida, a toda prisa, se acercó al rey y le pidió: «Quiero que ahora mismo me des en una bandeja la cabeza de Juan el Bautista». El rey se puso muy triste; pero por el juramento y los convidados no quiso desairarla. Enseguida le mandó a uno de su guardia que trajese la cabeza de Juan. Fue, lo decapitó en la cárcel, trajo la cabeza en una bandeja y se la entregó a la joven; la joven se la entregó a su madre. Al enterarse sus discípulos fueron a recoger el cadáver y lo pusieron en un sepulcro.________________Descárgate la app de SerCreyente en https://sercreyente.com/app/¿Conoces nuestra Oración Online? Más información en: https://sercreyente.com/oracion¿Quieres recibir cada día el Evangelio en tu whatsapp? Alta en: www.sercreyente.com/whatsappTambién puedes hacer tu donativo en https://sercreyente.com/ayudanos/Contacto: info@sercreyente.com

Razones válidas para que el nombre que eligieron tus padres para ti no sea de tu agradoECDQEMSD podcast episodio 6231 No Me Gusta Mi NombreConducen: El Pirata y El Sr. Lagartija https://canaltrans.comNoticias Del Mundo: Elecciones ticas - Laura Fernández presidenta de Costa Rica - La Bukele tica - Escándalo en los Gramys - Trump demanda a todos - Tio Richi se pone al día - Grupo Salinas y el SAT - Carnaval de Venecia - La ratotaHistorias Desintegradas: Nombre y sobrenombre - Modas y generaciones - Lápidas, jabones y personajes - La recomendación - Un halago inconsciente - Los buenos regalos - Creadores del cucú - Largo viaje - Unos mates - Día internacional sin pitillo, pajilla, popote, absorbentes, bombillita - El Golden Retriever - A los abogado - Elmo de Barrio Sésamo - Aniversario del INAH y más...En Caso De Que El Mundo Se Desintegre - Podcast no tiene publicidad, sponsors ni organizaciones que aporten para mantenerlo al aire. Solo el sistema cooperativo de los que aportan a través de las suscripciones hacen posible que todo esto siga siendo una realidad. Gracias Dragones Dorados!!NO AI: ECDQEMSD Podcast no utiliza ninguna inteligencia artificial de manera directa para su realización. Diseño, guionado, música, edición y voces son de  nuestra completa intervención humana.

El pasado viernes el departamento de Justicia liberó el segundo y definitivo lote de documentos relacionados con el caso de Jeffrey Epstein. El primero de ellos vio la luz en diciembre y este era muy esperado. En total se han liberado más de tres millones de páginas en las que también hay miles de vídeos y fotografías. Esto ofrece una perspectiva mucho más detallada de la vida de Epstein y de las privilegiadas relaciones que cultivó en vida. Eso sí, como sucedió hace algo más de un mes, esta nueva documentación también está plagada de tachaduras que el Gobierno justifica para proteger el anonimato de las víctimas. El fiscal general adjunto, Todd Blanche, ha defendido la integridad de todo el proceso. En una rueda de prensa que concedió nada más hacerse públicos los documentos aseguró que se había cumplido estrictamente con la ley y que no se estaba protegiendo a nadie, tampoco al presidente. Pero, como era de esperar, las críticas no han tardado en arreciar. Unos creen que las ediciones sobre los textos y las imágenes son demasiado agresivas y ocultan importante material que debería ser público. Otros denuncian que se ha revelado indebidamente la identidad de al menos 43 víctimas. El abogado de estas últimas, Brad Edwards, ha señalado que el departamento de Justicia conocía estos nombres de antemano, lo que podría acarrear consecuencias negativas para quienes deseaban mantener su anonimato. Entre los nombres que han vuelto al primer plano destaca el de Donald Trump. Los archivos contienen miles de menciones al actual presidente, algo que tampoco es extraño ya que es del conocimiento público que Epstein y Trump fueron amigos en los años 90. Aunque Trump insiste en que cortó toda la relación con Epstein antes de su primer arresto, la publicación de correos de Melania Trump elogiando a Ghislaine Maxwell y registros de vuelos antiguos mantienen la mirada fija sobre el inquilino de la Casa Blanca. Hasta el momento no hay pruebas de que el presidente participase en delitos sexuales. De hecho, él mismo se ha apresurado a recordar que estos documentos le absuelven de cualquier sospecha. El mundo empresarial también se ha visto salpicado. Los documentos revelan correos electrónicos con Elon Musk en los que ambos coordinaban posibles visitas a la isla de Epstein entre 2012 y 2014. Musk ha negado rotundamente haber asistido a ninguna de sus fiestas. Por su parte, Bill Gates aparece mencionado en correos donde Epstein aludía a sus problemas matrimoniales. Richard Branson, fundador de Virgin, y Serguéi Brin, fundador de Google, también aparecen en la documentación, en ambos casos por relaciones personales o de trabajo en el pasado. Brin, por ejemplo, recibió asesoría fiscal de Epstein hace casi 20 años. Por último, el caso del ex príncipe Andrés de Inglaterra continúa empeorando. En el nuevo lote de documentos aparecen fotografías suyas con una joven tendida en el suelo y la prueba de que invitó a Epstein al Palacio de Buckingham en 2010. A medida que se analizan estos millones de páginas, queda claro que, aunque la presencia en los archivos no implica necesariamente la comisión de un delito, la sombra de Epstein sigue generando una profunda incomodidad en las esferas más influyentes del poder político y económico mundial. En La ContraRéplica: 0:00 Introducción 3:55 Epstein 33:43 “Contra el pesimismo”… https://amzn.to/4m1RX2R 37:51 La caída del dólar 43:41 El hallazgo de Barbacid 47:59 Elecciones en Costa Rica · Canal de Telegram: https://t.me/lacontracronica · “Contra el pesimismo”… https://amzn.to/4m1RX2R · “Hispanos. Breve historia de los pueblos de habla hispana”… https://amzn.to/428js1G · “La ContraHistoria del comunismo”… https://amzn.to/39QP2KE · “La ContraHistoria de España. Auge, caída y vuelta a empezar de un país en 28 episodios”… https://amzn.to/3kXcZ6i · “Contra la Revolución Francesa”… https://amzn.to/4aF0LpZ · “Lutero, Calvino y Trento, la Reforma que no fue”… https://amzn.to/3shKOlK Apoya La Contra en: · Patreon... https://www.patreon.com/diazvillanueva · iVoox... https://www.ivoox.com/podcast-contracronica_sq_f1267769_1.html · Paypal... https://www.paypal.me/diazvillanueva Sígueme en: · Web... https://diazvillanueva.com · Twitter... https://twitter.com/diazvillanueva · Facebook... https://www.facebook.com/fernandodiazvillanueva1/ · Instagram... https://www.instagram.com/diazvillanueva · Linkedin… https://www.linkedin.com/in/fernando-d%C3%ADaz-villanueva-7303865/ · Flickr... https://www.flickr.com/photos/147276463@N05/?/ · Pinterest... https://www.pinterest.com/fernandodiazvillanueva Encuentra mis libros en: · Amazon... https://www.amazon.es/Fernando-Diaz-Villanueva/e/B00J2ASBXM #FernandoDiazVillanueva #epstein Escucha el episodio completo en la app de iVoox, o descubre todo el catálogo de iVoox Originals

Bad Bunny hace historia en los Grammy y habla en favor de los inmigrantes: "No somos animales". Liam Conejo y su padre ya están libres y en Minneapolis. Unos 137 millones de personas están bajo alerta por el frío extremo, que es récord hasta en Miami.See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.

Bienvenida FAMILIA NOCTURNA a este especial del Millón en ⁨@HABLEMOSDELOQUENOEXISTE⁩ YA SOMOS 1 Millón en la Familia y contando.Hoy quiero contarles historias aterradoras que suceden en rancherías donde un oficial enfrentará el horror mas grande de su vida, un enfermero que atenderá su último paciente. Unos hermanos que encontraron un lazo muy profundo, casi tan profundo como la pesadilla que vivíanY esto es, solo el inicio de este especialPor un millón de suscriptores.Gracias Familia!!!ORGULLOSAMENTE FAMILIA NOCTURNAVAMOS POR ESTE 2026!!! ⁨@HABLEMOSDELOQUENOEXISTE⁩ es un canal con el formato podcast que comenzó en abril del 2022, su primer episodio fue "vivo en un casa embrujada" en el que una chica narró sus vivencias y sucesos paranormales a lo largo de 20 años en la casa de sus padres, desde ese episodio hablemos de lo que no existe ha marcado una tendencia en exponer casos paranormales de personas comunes que viven en diferentes partes del mundo. Ice Murdock es el conductor o host de éste canal, durante casi 100 episodios no apareció, nadie conoció su rostro y la comunidad de éste canal , la familia nocturna , creó teorías acerca de quién era el dueño de esa voz. Hablemos de lo que no existe se destaca por tener apertura ante las opiniones experiencias y vivencias de cada uno de los invitados. La comunidad de éste canal es conocida como la familia nocturna, de hecho por estar leyendo o escuchando esto, tu ya eres miembro de la familia nocturna.. bienvenido. El duelo de historias es un concepto que se creó en el canal @Hablemosdeloquenoexiste, idea original del narrador, se estrenó en el episodio "Comité de la Muerte ,historias de Hospitales" el 1 de junio de 2023 y empezó a implementarse formalmente en el episodio "Abrí la puerta a un Demonio" el 11 de Enero de 2024 ; consiste en un duelo entre Narradores, una dinámica sencilla, donde cada uno cuenta una historia y busca superar a la anterior y al final la familia nocturna nos comparte en comentarios cual fue la historia más aterradora. El Narrador y todo el equipo de Hablemos de lo que no existe trabajamos para darles a ustedes querida Familia nocturna contenido original y de calidad, tardamos a veces semanas ideando formatos luego de tomar en cuenta las cosas que nos han pedido a lo largo de la temporada anterior y por eso el día 23 de Septiembre de 2024 comenzamos una nueva temporada que llamamos FOGATA DE HISTORIAS, en donde el narrador prepara una serie de historias escalofriantes una tras otra para retar al espectador a terminar el episodio por el nivel de miedo que genera. En este canal se relatan historias de terror paranormales, sobrenaturales y reales, prepárate para conocer el miedo de una forma en la que nunca lo habías experimentado.​

Al cabo de una intensa andadura por innumerables agrupaciones habaneras, iniciada en la frontera de los años 30 a los 40, donde destacaron la orquesta "Bellamar" de Armando Romeu y la orquesta de René Touzet, el trombonista, arreglista y compositor guantanamero Leopoldo "Pucho" Escalante logró canalizar su pasión por la improvisación y la descarga con la fundación del llamado "Noneto Cubano de Jazz" en el mismo centro de la convulsa década del 60. La aparición del noneto hacia febrero de 1964, bajo la batuta de este destacado músico, con una sonoridad y rítmica enmarcadas básicamente en la corriente del swing, aportó una mayor consistencia a la categoría del jazz en el espectro discográfico cubano. Así recordamos al trombonista Pucho Escalante, uno de los grandes pioneros del jazz que es imprescindible tener en cuenta. A punto de arribar a los 102 años falleció en Nueva York el 17 de octubre de 2021. Había nacido en Yateras, Guantánamo, el 14 de diciembre de 1919. Unos minutos en la criollisima compañía de María Cervantes. Notable pianista, compositora e intérprete, su incesante paso por la radio, las salas teatrales, clubes y cabarets, entre otros escenarios, dejó una impronta de auténtica cubanía. De su padre, el ilustre Ignacio Cervantes, recibió las primeras clases de piano, así como todo un universo melódico que, aparejado al respeto a la música y al público, le acompañó hasta el final de sus días. Gracias a las grabaciones discograficas que registró para la etiqueta Columbia, a partir de los años 20, su arte comenzó a trascender nuestras costas. María abrió un singular camino que tuvo en Bola de Nieve el más alto exponente. Nació en La Habana el 30 de noviembre de 1885. Falleció a los 95 años el 8 de febrero de 1981. Y a propósito de María Cervantes llega a nuestra memoria el Bola. Combinó de manera genial sus grandes cualidades al piano con una originalidad interpretativa aún rara de encontrar en estos tiempos. Con absoluta humildad, siempre fiel a su arte, con mucho sentimiento, hizo suya la canción. Desde el Hotel Internacional de Varadero, como parte del Festival de la Canción del año 1970, llega el inigualable piano man. Probablemente una de sus últimas presentaciones, teniendo en cuenta que fallecería en México un año más tarde, el 2 de octubre de 1971 Casi en la despedida revisitamos el panteón de los pioneros del jazz cubano. La antigua señal de la CMQ de Monte y Prado de finales de 1945, nos recuerda el estilo de Dandy Crawford. Como elemento indispensable en el elenco que proponía el show "El club del Swing", Dandy llamaba la atención con el apoyo de la orquesta CMQ bajo las conducciones de los maestros Alfredo Brito, Armando Romeu y Félix Guerrero. Por entonces el swing y el be bop cautivaban a músicos y bailadores.

Hoy debatimos sobre el apasionante tema de cuando casi se acaba el jabón pero todavía un tiro más de vida útil. APASIONANTE. Unos ravioles de ricota - Miércoles 28 de Enero, 2026

Nieves Concostrina habla sobre la fundación de Sídney como una colonia penal británica debido a la saturación de las cárceles en el Reino Unido.

La vez pasada hablamos acerca del llamado a entender cual es la voluntad de Dios, es decir, Su voluntad revelada aplicando los principios bíblicos a nuestras situaciones cotidianas aunque no necesariamente haya un mandamiento específico. También hablamos del llamado a evitar el desenfreno que produce lo que nos intoxica mas bien necesitamos la claridad que produce la llenura del Espíritu Santo la cual sucede cuando oramos y creemos la Palabra; por eso debemos alimentar nuestra fe dándole prioridad a las Escrituras en nuestras vidas diariamente.

Unos 185 millones de personas siguen bajo alerta invernal después de un fin de semana gélido. Más de 20 estados declararon estado de emergencia por el frío extremo. Nuevos videos de la muerte de Alex Pretti a manos de federales contradicen al Gobierno.See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.

Nieves Concostrina habla sobre la fundación de Sídney como una colonia penal británica debido a la saturación de las cárceles en el Reino Unido.