Podcasts about PY

We're announcing AIEWF speakers this week! Take the AI Engineering Survey!Today's guest Ethan first joined us for the LS Paper Club as the lead on NVIDIA Cosmos World Model, but then joined xAI and built Grok Imagine in 3 months:He comes back on Latent Space with some nuclear hot takes: that Video Models primarily get their intelligence from LLMs, not from training on video data, and that the next frontier for truly interactive, realtime, long-horizon world models is to work on LLMs (perhaps Interaction Models as well…)Put it this way: In the near term, the next Sora won't be a better video model, but a video agent.Generative Media may more closely follow the evolution of AI coding which went from focusing on one-shot output performance and cost, to multiturn reasoning and planning models for agents and systems that can plan, edit, test, debug, and submit PRs.At a certain point, coding models got so good that the only significant next step to improve performance was handling the orchestration of these models.Now as the performance of video models increases significantly across realism, consistency, & prompt adherence while becoming more cost efficient, the next evolution of video generation may also be systems that can plan, generate, edit, critique, and iterate across an entire creative task. In this episode, Ethan joins swyx and Vibhu to unpack what it actually takes to build frontier image and video systems: data, VAEs, diffusion transformers, audio-video alignment, inference speedups, and the hidden cost of storing and moving massive video datasets. From building NVIDIA's Cosmos world model to joining xAI as Grok Imagine was being built from zero to one, Ethan He has been at the center of some of the most important work in video generation, multimodal models, and real-time world models.We go deep on Grok Imagine, how a small xAI team shipped its first multimodal video model in three months, why iteration speed matters more than almost anything in model development, and why many of the biggest gains come from fixing tiny bugs in data and training pipelines. Flipbook: The future of VideomaxxingVideo agents are almost a sure bet to be the trend in the coming year. We end with a glance at what's beyond video agents:Flipbook caused a minor sensation this year when it was released, but most treat it as a fun demo. Ethan takes it very seriously — with the speed and cost of inference coming down every year, the future of custom video JIT UI is closer than you think. We talked about why videogen models may become the front end of AI, how generative UI could replace traditional HTML/CSS, why world models need to be real-time, interactive, and long-horizon, and why the future of video generation may depend more on language models and agents than on diffusion alone.We discuss:* Why fast iteration mattered more than meetings* Why small training bugs can drive huge model quality gains* Why coding models may make compute the bottleneck again* How image and video models are trained with synthetic captions* The role of VAEs and latent space in frontier video models* Why image models are the foundation for video models* The tradeoff between temporal compression and real-time interactivity* Flipbook, Neural OS, and the future of generative UI* Why future interfaces may go from user intent to pixels* The hidden cost of training video models: storage, egress, and GPU hours* How step distillation and consistency models (like OpenAI sCM) makes video inference orders of magnitude faster* Grok Imagine 0.9 and large-scale audio-video generation* Why audio-video alignment is harder than text-video alignment* Ethan's definition of world models* Reference-to-video, video extension, and long-context video generation* Why xAI's research communication undersells Grok Imagine* How xAI culture shaped the speed of development* AI watermarking, SynthID, and detecting generated media* Why prompt rewriting matters for video models* Grok Imagine Agent and the rise of video agents* Why language models may unlock better video generation* Robotics, physical AI, and embodied world models* Why Ethan left xAI and shifted focus toward LLMs* Self-managed context, memory, and the next frontier for language modelsEthan He* LinkedIn: https://www.linkedin.com/in/ethanhe42* X: https://x.com/EthanHe_42Timestamps00:00:00 Introduction00:01:25 From NVIDIA Cosmos to xAI00:03:24 Building Grok Imagine from Zero to One00:10:07 How Image and Video Models Are Trained00:18:53 Video Compression, VAEs, and Real-Time Tradeoffs00:22:10 Generative UI, Flipbook, and Neural OS00:32:10 The Cost of Training Large Video Models00:37:04 Distillation, GANs, and Fast Video Inference00:41:21 Audio-Video Generation and Grok Imagine 0.900:48:34 What Makes a World Model?00:55:51 Reference Videos, Long Context, and Video Memory01:00:11 xAI Culture, Research, and First-Principles Building01:09:45 AI Safety, Watermarking, and Prompt Rewriting01:13:10 Video Agents and AI-Assisted Creation01:27:32 Why Language Models Unlock Better Video01:31:15 Robotics, Physical AI, and Embodied World Models01:32:38 Why Ethan Left xAI01:34:16 Self-Managed Context and the Future of LLMs01:38:43 Ethan's Career Path and Closing ThoughtsTranscriptIntroduction: Ethan He, Latent Space, and the Path to xAISwyx [00:00:00]: We're here in the studio with Ethan He, most recently of xAI. Welcome.Ethan [00:00:10]: Thank you. Glad being here.Swyx [00:00:11]: We're also here with Vibhu. you were first coming to us or joining the latent space world because you were working on Kosmos at NVIDIA, and you did a paper. We loved it. you presented it as well, so thank you for doing that.Ethan [00:00:23]: I've actually, I also presented the MoEs twice at latent space.Swyx [00:00:29]: How did you actually hear about us? Did we reach out to you? Is that how it worked?Ethan [00:00:33]: No, actually, I-- the community. Like I realized, oh, there is this online community that people talk about AI and also learn from each other through papers every week through the Paperclip. It's very nice.Ethan [00:00:49]: I learned a lot.Swyx [00:00:49]: I think three years stop. We haven't stopped even on Christmas and New Years. many weeks I want to stop but it keeps going.Vibhu [00:00:58]: No, that was good. I think you had posted that you worked on a paper, and I was “Oh, very cool. We have Paperclip. Present then.”Vibhu [00:01:04]: But I might have reached out to you after.Swyx [00:01:05]: you-- because it's an amateur club, right?Swyx [00:01:08]: so it's very unusual and but we have sometimes paper authors come by and actually explain the paper. Today we just did, the poolside paper, which was apparently very good.Vibhu [00:01:18]: Came out yesterday.Vibhu [00:01:19]: pretty interesting, right? Fully open. They talk about everything, systems. So it's a good one. We'll, we'll recommend people to read it.Swyx [00:01:25]: Bring us up to speed on your transition to xAI, ‘cause I actually don't even know when you joined. just like tell the, tell the story about the sort of transition.From NVIDIA Cosmos to xAI: Scaling Video and World ModelsEthan [00:01:34]: Before xAI, I was working on Kosmos world model as in-- at NVIDIA. So Kosmos is, it's a giant video foundation models that can-- that aims to simulate the world and for-- it serves as a foundation of-- for all of the roboticists to build on top of. There, once I built the Kosmos one, I realized as this thing also has a scaling law similar to language model, we need to scale up the video models further. that's, that's why I realized I need to move to somewhere with much more compute resources. That's how ISwyx [00:02:13]: Than NVIDIA?Vibhu [00:02:14]: The GPU rich came themselves.Vibhu [00:02:19]: And timeline-wise, when was Kosmo? It was pretty early, right? It was open world model, open paper, everything.Ethan [00:02:25]: It was end of twenty-four.Vibhu [00:02:28]: End of twenty-four.Ethan [00:02:30]: Then at mid twenty-five, I moved to xAI. At that time-- I joined about the time when xAI was about to build video models and in multi-model models. There were no infra, no data, and no model, and it just-- as a few engineers, we built it in three months and released the first model, Grok Imagine zero point nine.Ethan [00:02:55]: And since then, I keep working on video models and move more from training and to post-training of the video models. For example, like a reference to videos, kind of like the cameo feature and, video extensions. And, before I left, I worked on a world model, leading a small team to focus on the real-time long horizon video generation.Building Grok Imagine From Scratch in Three MonthsSwyx [00:03:24]: Can you give like a rough roadmap of okay, you're on a brand-new team. Grok previously was only text, or they partnered with BFL for their image gen stuff. What do you-- what are the building blocks, right? You have compute, data you can procure somewhere. Like just what are like the sequence of things that people should think about when you're setting up a new team?Vibhu [00:03:43]: actually even deeper, not just data you can procure. You guys had to go through getting the data too, right? So you shipped it pretty fast, but yeahSwyx [00:03:51]: three months is likeVibhu [00:03:52]: From everythingSwyx [00:03:52]: actually like very surprisingly fast.Ethan [00:03:55]: One thing I say like thanks to my experience at NVIDIA, ‘cause first time when we were building Kosmos together, we built it, for about a year. So this is like the second time I do it. Roughly have an idea, what to do. I say the most important thing is the talent. Everyone were very strong and clever, very close with each other towards a common goal. So that speed up things a lot. So you reduce the communication bandwidth among people, and everyone can work towards the same goal. It's, it's like every day there's not that much meetings on the calendar, like maybe like a, like a sync a day, and after that it's, it's just all building. It was pretty fun at that time.Ethan [00:04:47]: And another thing is that xAI has very strong foundations of like data inference, model inference, and the supporting there can help the model develop a lot. When I look at, training models, I don't so actually the top important thing is like how many, how many iterations can you do, per day? and the more iteration can you do, you can, you can train the model much faster. So if you have very strong infra and you have a lot of compute, you can, you can train these models in very short period of time. That can give you a much larger buffer to, for errors, and it also gives you the opportunity to spot more bugs.Iteration Speed, Compute, and Debugging Model PipelinesSwyx [00:05:46]: What is an iteration? Is it like a few hundred steps or what are youEthan [00:05:50]: Let's say just the train-training the model, like from acquire new data and maybe design new algorithms and train a new model, maybe at smaller scale orSwyx [00:06:01]: So cycle time for like any hyperparam that you're searching.Ethan [00:06:04]: Cycle time and tune to like eval this model. Is this model better than my previous iteration?Ethan [00:06:11]: SoSwyx [00:06:11]: So it's like before you, someone had already set this up that you can iterate very quickly.Ethan [00:06:15]: I think the foundation there is extremely good forDeveloping and research models.Ethan [00:06:23]: And often I find is it-- this is kind of boring, but like a lot of the improvements does not come from new algorithms. It comes from finding small bugs here and there in the data pipeline, in the, in the model training pipeline. Those give, those give the biggest boost to the model quality.Vibhu [00:06:46]: It's interesting, right? So you say it's like small team, less communication bandwidth, but also a lot of quality is like find little bugs. It seems counterintuitive, right? You have a lot of people, you can iron out more of those, but it's interesting to see the other side, right?Swyx [00:07:00]: I also wonder, have you-- do you try using LLMs to look for bugs? I don't know.Ethan [00:07:05]: I remember at that time it was mid two thousand and twenty-five, so it's the coding model wasn't quite there yet. I remem- I remember like December two thousand and twenty-five, it was extremely good. Yeah, I've been, I've been using it at that time. It's, it's helpful. sometimes it produce codes that are kind of difficult to maintain, even though like the first time it built something extremely fast. But it gave the, like a spaghetti code, thousands of lines that I couldn't maintain, and the LLM itself couldn't figure out what's, what's wrong and how to improve on top of it. But now I find it much better. Yeah, I want to bring up another point here is now coding models are much more efficient and can help us implement stuff much faster. Compute might become a bottleneck again because previously, like if you want to train a new model, say you want to generate new synthetic data and then or write a new algorithm, it might take a few weeks. And during that period of time, you don't-- you might not have experiments to run. But now you can build that thing within a few hours, then you can immediately train a model.Ethan [00:08:24]: Now you have to have enough compute to try all of the ideas. So compute might be the bottleneck of iterating speed again.Swyx [00:08:36]: yeah, I actually, honestly, I think it's like kind of a stressful job because you're “Well, I should be trying everything, and if I'm not, then I'm not doing my job well.”Vibhu [00:08:48]: there's also the stress of you're eating thousands of GPUs per hour, which is very expensive and, compute can go to other researchers.Swyx [00:08:56]: You got the daddy Elon toVibhu [00:08:57]: You got daddy Elon.Ethan [00:08:59]: It wasVibhu [00:09:00]: But there's still finite amount of compute, like you want to use it, you want to use it well, you want more of it.Ethan [00:09:06]: That was quite stressful indeed. Yeah, I think one thing is the-- with coding models now, like a lot of these jobs can be automated, which is much better. A second, it's a, it's a marathon, so you got to maintain good health and, a regular schedule.Vibhu [00:09:28]: It's, it's hard to hear that when you shift from zero to nothing in two months.Swyx [00:09:32]: and, I think obviously the culture at xAI is very famously, people work very hard. one thing I did want to dive into, in our-- in the notes that you, that you sent ahead of time, you had specific comments about the cost of Video Gen training. presumably this is on the Colossus-1, right? the two hundred megawatt cluster. Any whatever you want to just share on that.Vibhu [00:09:54]: I think there's, there's three things we're talking about, right? So there's Video Gen, there's also the Image Gen model that you put out. Do you want to like complete the, okay, so zero to one, you have a few months. Just what are the stages of create Image Gen model?Swyx [00:10:06]: Oh, yeah, maybe I got distracted.How Image and Video Models Are Trained: Synthetic Captions, Tokenizers, and VAEsVibhu [00:10:07]: Sorry. and then, from there's Video Gen, there's Audio Gen. Would love to get into those next. But what is that first few months like? So small team, a lot of bugs, iterations, but what does it look like? Do we take something off the shelf? Do we just get data compute? What's, what's the few months like? How do you go to state-art Image Gen model? How do you just start?Ethan [00:10:28]: I cannot comment specifically how xAI did, but it's, it's a quite standard process. I can draw some, examples from Cosmos. So mainly it's building a video model, you actually need to build a image model first. And building these two models, the data you need is a hundred percent synthetic pair of language and image or language to video. Because on the, on the internet, actually, the videos don't naturally associate with text. So you can say, oh, like on YouTube, you have the title and you have the description and the commentsSwyx [00:11:11]: TitleEthan [00:11:11]: of a video, but usually they're not relevant to the video itself. And say maybe like the video is a natural scene of mountains or something, and the title is, I'm so happy today.Ethan [00:11:26]: So they have they have no correlation at all. So the first step is to, you have to generate synthetic pair of language with the videos. So you gather videos from the internet, and you use a VLM to caption the videos. So that part, here's a question, like how do you, how do you gather VLM to begin with? So if there's noSwyx [00:11:55]: You, so you fuse the model, right? LikeEthan [00:11:57]: Say if there's no like VLM exists, like how do you generate the text to the beginning, right? It's, it's impossible.Swyx [00:12:04]: I see.Ethan [00:12:05]: In the beginning, it's like you ask human to describe the video as detailed as possible.For example, you ask them to describe everything, like all objects, all characters, and all interaction and dialogues in the, in the videos. So that's in the protocol of Cosmos labeling. We require the objective we give to the labelers was that you have to describe the video as detailed as possible, such that a blind person hears a blob of text can reconstruct what the video is like from their head.Swyx [00:12:43]: Video or image? You're talking about images.Ethan [00:12:44]: Video or image, either one of them.Vibhu [00:12:47]: This was pretty common when we went from clip and DALL-E, right?Vibhu [00:12:51]: It's all training on really detailed captioning of images. So same is applied to video, but insteadEthan [00:12:57]: same appliedVibhu [00:12:57]: of using multimodal model to pass in video images and write rich descriptions, you can alsoSwyx [00:13:04]: I think there's this traditional perspective of supervised, or, very highly human curated thing. I feel like there's a unlock with unsupervised, right? Where like you have enough to bootstrap that you can just throw common corpus on it or, whatever. like unsupervised vision and language pairing, right? Like where you just have, interspersed image and text and it just learns. To me, that is the VLM breakthrough that is different from the clip, different from the LM era.Ethan [00:13:36]: It's interesting to see that you kind of need both data.Ethan [00:13:41]: For example, for theSwyx [00:13:41]: You need it to bootstrap it up. YeahEthan [00:13:43]: for the generative model training, there's also usually like a small percentage of unlabeled data. So the model is instructed to generate a video without any text instruction. That can also help the model generalize. So after this stage of generative synthetic pair, so, one important common step is to train a compressor or a tokenizer of the image or videos. So because, if you train-- If you can technically, theoretically train image or video models on pure pixels, but the problem is that the, it's, it's a lot of tokens. So like one image, it's, a thousand by a thousand, it's like one million tokens, one million pixels. It's impossible to train transformer on that. So it's, you need to train a tokenizer, which can go from image to latent space and latent space back to image.Swyx [00:14:45]: That's why we named the podcast.Swyx [00:14:48]: But, basically, you're talking about vocabulary science.Ethan [00:14:50]: so vocab.Swyx [00:14:51]: And so, what is, what is imp-- like a million is impossible?Ethan [00:14:54]: In generative models, the vocab is continuous. It's a continuous space. We can think about like you map an image to a vector. It's a, it's a fixed length vector. It's sixteen or forty-eight, something like that. And then you map that vector back to the image space. And the mapping is, has-- The mapping is patch-based. So you say you haveEthan [00:15:22]: a sixteen by sixteen patch and you match, you map that patch of pixels into this latent space.Swyx [00:15:29]: We've covered thisVibhu [00:15:30]: This is like the vision transformersSwyx [00:15:32]: VAEs,Ethan [00:15:33]: VAEs.Vibhu [00:15:34]: You basically compress your input, you do your generation, you're reasoning all that generation in smaller dimension, and then you project back out.Swyx [00:15:43]: VAE is a form compression, but I think the for me, the patching thing is from VIT, right?Ethan [00:15:48]: You can make those.Swyx [00:15:49]: Literally the, yeah, the paper is titled like sixteen by sixteen is all you need. something like that. and then I think also, people make a lot of comparisons with this kind of patching with convolutions.Swyx [00:16:02]: Which is you're, you're kind of re- reconstructing the old paradigm with the new.Ethan [00:16:05]: Actually, in VAEs, there are, there are both convolution networks and transformers. You can actually do both.Ethan [00:16:14]: After this VAE, so what you've got is you've got latent space tokens and you've got the language tokens. So now the training of the diffusion transformer, usually generative models use diffusion transformers. It is actually quite standard. It's, it's very similar to how you train a language transformer models. It's not that much difference. It's just the tokens, the visual tokens in, visual tokens out. The only difference is there's a denoising process. So you train the model to unmask some of the noise. So you add, you add random noise to the visual tokens, and then you train the model to remove those noise to generate the clean tokens. Any inference, the model can iteratively remove noise from a hundred percent noise.Swyx [00:17:12]: And then there's also, to speed things along on the tech tree of diffusion, there's CFG, and then there's, there's also, latent diffusion that, there's, there's someone in there. I think, somewhere along the line, obviously, like stability and all these other guys, pioneered a lot of this, architecture. I don't know if you want to get into that or just, or do the video side up to you.Bootstrapping Video from Image Models and Temporal CompressionEthan [00:17:37]: After you train such model, such image model, the reason it's a, it's a foundation for video models is that image models are cheaper to train, and they have much denser connection between language and text. So, sorry, language and images. For example, you train a billion, you train on a billion images, and there's a mapping from the text to the image. And the cost to train the same, like the, a billion, a billion text to a billion videos, that's much more expensive because videosNaturally have more tokens than images. Because the diffusion models, their understanding of, language purely come from this mapping. So if you don't have enough mapping, so if you only train on like a ten million videos or something, there-- you might not see enough language tokens in your training, so your model does not understand human intention enough. So that's why you really-- you train-- you first train this image diffusion models, and then you bootstrap the video model from there.Swyx [00:18:53]: One thing I did want to ask, because I-- actually, I think you're, you're the first per-- video model person I've ever talked to, I think. we've, we've like talked to Luma and all those folks. There's all these tricks in video compression where basically frame by frame there's not that much difference, so actually you don't have to regenerate or save the whole frame, right? but I think MP4 compression or something else like that.Swyx [00:19:16]: is it tempting to use that? Or as far as I can tell, everyone just treats it as, “No, we would just generate every frame.” Is that roughly the state-art?Ethan [00:19:27]: There are a few different approaches. Let's say first, like you want to just directly use MP4 compression and use that as the tokens for the transformers to train, right? So people actually have tried that, but the main challenge is the latent space for the MP4 tokens were not, were not very comprehensible for the models. It's, it's extremely hard to train on that. And there's aEthan [00:20:01]: So that's why they created VAEs, which creates more continuous, latent space, so the models can understand that latent space and learn from it much easier. Even within the VAEs, there are different difficulties of the latent space. So you can imagine something the simplest, the most naive VAE is like you have an image, and you just shuffle all of the images into a, into a vector. So you don't need to train any VAEs, right? But that latent space is extremely hard for models to train on top of. That's why there are some debate on like how do you compress the tokens. So you mentioned like you can compress frame by frame. Also, you can compress, the temporal dimension.Ethan [00:20:52]: The difference is if you compress the temporal dimension, you get a much higher compression rate. Because there's temporal redundancy between frames, because, this frame and the last frame, likely they are mostly similar, so there's only some small difference. for example, I think in 12.1 VAE, they have like a eight by eight by four compression rate. So the four temporal tokens are compressed into one tokens. That can save a lot of, save a lot of the context length. If you do it frame by frame, you have to do maybe like eight by eight by one. Your context length will be four times larger. That being said, the benefit of the frame-- per frame compression, we might come back to this later, is, real-timeness and interactivity. ‘Cause if you, if you strain the output of the model, frame by frame, you can-- the model can respond to any user request immediately. So if you have like a temporal four compression, four times compression, thenSwyx [00:22:06]: It might be laggyEthan [00:22:07]: there's a lag there in nature.Swyx [00:22:10]: So you're very pilled on this. let's just go ahead and bring it up ‘cause we have the visual prepared anyway. There's some frontier applications of real-time video gen. So Flipbook is one of the examples that went viral recently, right? What is Flipbook?Real-Time Generative UI: Flipbook, Neural OS, and Diffusion Front EndsEthan [00:22:23]: Flipbook is kind of like a web brow- web browser. You can see like it has the web bro- browser UI on top. The difference is all of the UIs are generated by generative image model in real time, and anything here are fake. But you can, you can explore inside this wor- this imaginary world. Say like we-- here we have engineering the Great Pyramid. Like the model generates this for us to understand how it works, and if we want to navigate around and understand further, we can click on some of the, some of the description here, and the model will generate a new page, new subpage describing the details we want to know about.Swyx [00:23:14]: So it's basically kind of we're playing a video, but it's pausing for our next interaction, and then it just plays the next thing based on our interaction.Swyx [00:23:23]: Which is kind of cool.Vibhu [00:23:25]: and you kind of decide your story. So this was, how do you make a pyramid? levering technique seemed interesting, right? It shows how do you take Okay, I want to know what is thisSwyx [00:23:35]: The demo, the demo tweet had more animation between frames.Vibhu [00:23:38]: I think it's just skipping,Swyx [00:23:39]: Oh, it's just skipping a lot of frames.Ethan [00:23:40]: they also have a video modeVibhu [00:23:42]: It takes a lot. There's a lot of peopleEthan [00:23:42]: but, a lot of people are using it.Ethan [00:23:45]: So it's not available.Vibhu [00:23:46]: There's a live video stream. We can try,Swyx [00:23:50]: So this is an example of the kind of future that you see at the extreme. We don't-- we're obviously not in it today.Swyx [00:23:56]: But in a world where inference is completely free this is better than generating code and text?Ethan [00:24:02]: So this is, this is a final state of where Viva will be at for word model, I think. Imagine internet doesn't exist, and then you type in google.com. Like what should, what should, what should a model show you?the model can imagine something, and this is what the model imagine. And these web pages, they completely do not exist. So I think as the inference costs come down, we are going to have generative UI for everything. If you think about how the coding model works, so they write code for a web page, and they render the code might be con- converted into binary, and the binary render the pixels on the screen. So we in machine learning, every time we have some breakthrough, obviously it's, it's more intuit. So why don't we have like user instruction to the pixel directly? So the generative UI will be user intention to the pixels directly. And say like even if I want email, let's say everyone have the same interface, but I want, I want it slightly different. I want the email to show to me like a TikTok, so I can swipe left and right for the emails. And or maybe you want something else. We can have completely different things. Or like I have I'm looking at, Instagram stories, and I don't like the Like button. I always may click it. And, generative UI resolved it. So it's going to be a revolutionary replacement of the interface. So in the future, we might have much more powerfulEthan [00:25:50]: LLMs and coding models running behind the scene. And in the, in the front-end, the diffusion model will actually be the front-end to show stuff to you. That's how I imagine it.Swyx [00:26:02]: Diffusion front-end, deterministic back-end.Swyx [00:26:04]: Something like that. I find that very expensive, but,Vibhu [00:26:08]: I find it interesting you called LLMs writing code on the back end deterministic, but okay.Swyx [00:26:14]: you write it onceVibhu [00:26:15]: Compare it toSwyx [00:26:16]: And then you execute.Ethan [00:26:17]: If you think about the cost, say, let's say H100 costs $1 per hour, and if you use this eight hours a day and thirty days, so, every month you're paying this two forty, you'll actually not wanna pay for that. That's even more expensive than Cloud Code Max. But if you think about the compute costs come down like two times every year, and I think the future will likely arrive like within few years.Vibhu [00:26:49]: It's everything, right? compute cost comes down, compute gets faster, model gets smarterEthan [00:26:54]: More efficientVibhu [00:26:54]: model gets smaller.Swyx [00:26:55]: I don't know why you say two times, ‘cause I think it's like 100 times. In language models, it is roughly one hundred to a thousand times every twelve to eighteen months, for the same given level of LMSys, ELO.Vibhu [00:27:08]: That's a net of everything, right? That's model performance alongside compute. So different than just compute costs come down. But, a very interesting future.Swyx [00:27:19]: So the web designers will have to shout out that accessibility is an issue, right? how do you deal with screen readers or whatever. But yes, this is higher bandwidth storytelling than anything you can possibly generate with code, right? So I think that's the rough idea.Ethan [00:27:34]: And I'd like to add a little bit that so human naturally have the maximum bandwidth when we are looking at things, look at videos, and we also have maximum output bandwidth when we are talking. So in the future, it might be something like we talk to AI models, and the AI model responds back with a generative UI. So that would be the maximum input and output bandwidth to interact with AI models before neural link happens.Vibhu [00:28:06]: And it's also very custom, right? Some people are very visual, some people are not as visual, right? They prefer the text. But the best thing about generative UI, right, it can also be text.Swyx [00:28:17]: There's another project that we wanted to highlight, which is the Neural OS. Kinda similar idea, but here you're literally operating, simulating an operating system with a video model.Swyx [00:28:27]: and you can play Doom, you can do Firefox. I find this like mildly less impressive, obviously, because it's an OS that I can run.Swyx [00:28:37]: But here everything is imagined.Vibhu [00:28:40]: I was, used to the Command+W to close the Firefox tab. It didn't crash. That's why I saidSwyx [00:28:45]: It's too immersive.Vibhu [00:28:46]: It's, it's too immersive for me.Swyx [00:28:47]: Too immersive.Vibhu [00:28:48]: I wanted to close the tab.Vibhu [00:28:49]: But yes, I can play generated diffusion.Swyx [00:28:51]: this is shockingly fast.Swyx [00:28:54]: Because I remember there was a demo about like maybe one to two years ago. Someone tried to do the first-person shooter with a image model. There was no consistency. It was very slow. But here it looks like realistically it's-- this is Doom.Vibhu [00:29:07]: I think there's two sides to that, right? There's okay, what is running a game? The heavy part of it is actually the game engine, all the lighting, all that stuff, the graphics. This is just kind of video, right? Like we've solved consistency. This is still, it looks like a few years old image generation. There's some temporal consistency, but it's, it's kind of just images stitched together as frame video. But it's a good visual representation to pi- to picture the future you wanna see, right? that's, that's what I see in these more so.Ethan [00:29:38]: This reminds me of how the video models gets better and better. So Neural OS is kinda if you just look at it feels like it's just a crappy version of the, like the Windows we could have, right? And, but the difference is, so the model, this model is overfitted on the existing operating systems. It can generate nothing different than that. But it's actually also similar to video models. So when we are training these video model, image model, we train them on internet. There's no imaginary supernatural stuff on the internet. But once we train this model, you can prompt the model to generate something supernatural that have never existed in the data set. So if you train your Neural OS or neural computer on the standard screen recordings on the entire internet. The model can imagine completely new interface to interact with the computer.Swyx [00:30:43]: This is one of those things that is magical to me. usually generalizing out of distribution is bad, but somehow we have learned some kind of internal world model that you say, this plus, but it looks like rainbows and butterflies, it'll do it and it will kind of make sense.Swyx [00:31:03]: So yeah, that's kind of cool. Yeah, I don't know if there's any comment more on there. I do, I do wanted to, I did wanted to touch a little bit more on the model architecture stuff, which I think you were getting. It's, really fascinating. We don't get a chance to talk about this enough. So one of the papers that we covered, we've covered every annual, segment anything release. and I don't know if you follow-- you're a computer vision guy, so youEthan [00:31:26]: I knowSwyx [00:31:27]: . So they did memory attention, which is kind of interesting. And I always think, anything where you can, across the temporal dimension, keep some consistency, I think it's, very fascinating, and I don't know if Basically, does that-- the CV side bleeding into video gen side, I think is underexplored, right? we talk about it for labeling, but actually you can borrow the architecture itself.Ethan [00:31:50]: There's, there's also complete different approaches, right? you brought up the term world model, so we went from video model to world model. There is diffusion, but there's also other approaches that people are doing. So maybe we get into those after as well,?Swyx [00:32:03]: He has a whole definition of world models and stuff. I feel like we threw a lot at you. Whatever you want to comment on.Why Video Models Are Expensive: Storage, I/O, and Training ScaleEthan [00:32:10]: I think one thing that we should actually comment back on is okay, so we were talking about the steps to train image gen to video model. One thing we don't see as much of is okay, you brought up the delta in training data, right? SoEthan [00:32:24]: you won't have as much a video model might not generalize, but what is the cost of training a large video model? So we know for LLMs roughly, okay, even like the poolside thing that came out today, right? It's a Gemma level model trained on roughly forty trillion tokens at this many H200s over this much time, right? You can see what is the exact cost of that. So how many GPU hours over how much H200 costs? So how do we do the back-end math of, same thing for video models, image models. How do you, how do you kind of break that down? I can share some back-envelope calculation. So surprisingly, video models is-- the cost is very-- is comparable to language models and obviously the largest scale is language model, maybe like a medium scale to language models. I said just storing the videos alone, it costs a lot. You can, you can maybe look up on AWS or something.Ethan [00:33:20]: You really, say if you have a billion videos and let's say, let's just say like each video, like five megabyte, then you need five petabyte to just store those videos. And also remember we talk about you use a VAE to compress the videos, and you also need to store, typically you need to store those continuous feature, in-- also in your storage. That's also comparable size with the videos themselves. So just storing these videos and the features is tens of petabytes alone. And,Swyx [00:33:58]: I just, I just looked up the calculation. Five petabytes on S3 Standard is one hundred K per month.Ethan [00:34:05]: AndSwyx [00:34:05]: It's comparableEthan [00:34:05]: and you needSwyx [00:34:06]: AndEthan [00:34:06]: And then like tens of petabytes, two hundred K. And even more expensive is you have the ingress and egress.Swyx [00:34:13]: Oh, yeah.Ethan [00:34:14]: Like you-- through the internet. You have to just to download those videos, I believe it's, it's more expensive on AWS than just storing those videos.Swyx [00:34:25]: Storing, yeah.Ethan [00:34:25]: And each training runs, you probably need to pull them once. If you train multiple times, it's, it's even more than that. So it's like just storing the network, those costs is just, it would be a few, a few millions per month to just storing everything, not to mention the GPU cost.Ethan [00:34:45]: AndSwyx [00:34:45]: my side tangent, the compute rental, like GPU rental is very efficient. There's one side, okay, you can be XAI and build your data center. Should we not just build our, storage compute as well? LikeEthan [00:34:57]: Of courseSwyx [00:34:57]: cloud cost compared to just,Ethan [00:34:59]: You save so muchSwyx [00:35:00]: store. Yeah, exactly.Swyx [00:35:01]: Especially with like egress and stuff. So.Ethan [00:35:04]: That's a good idea, but it also comes to-- there are some of its own challenges.Swyx [00:35:09]: Of course, of course.Ethan [00:35:10]: like people who build the GPU data centers, they might not expect this much, storage. And yeah, people build storage, typically they just build it somewhere with just CPUs.Swyx [00:35:23]: I just looked it up. Five-- AWS only charges for egress, not ingress. Tier five for five petabytes is two hundred and thirty K.Ethan [00:35:32]: Even more expensive than the storage.Swyx [00:35:34]: But storing is per month, right? You check in, then you cannot check out. so it's so cool. It's okay. So there's that side.Ethan [00:35:41]: So the TLDR, my backhand mathSwyx [00:35:42]: Data is larger than you think. Yes.Ethan [00:35:44]: my backhand math of GPU hours times GPU cost is also very much, I'm missing some storage.Swyx [00:35:49]: You're also-- you're basically like also more IO bound than normal training.Swyx [00:35:55]: Yes. ‘Cause like data loading, so caching everything, it becomes super important.Ethan [00:36:00]: So in Cosmos, we did a lot of optimizations to make it not IO bound. So, speaking of the training, actually training the model, the GPU cost, if you look up like the open source model, how big these video models are, I think like LTX has nineteen B parameters. That's a dense model. And people are also exploring, MoEs, so it might be twenty B active and, like a hun- hundreds B, total. So that's, that's even-- that's similar size as medium-sized LLM models. And if you, if you look at number of tokens-Uh, we disclose that in Cosmos. It's also like tens of trillions of tokens on the visual tokens. So putting this together, the cost of, training these video models, it's actually comparable with LLMs. Not to mention, the infra is slightly different from LLM, so it might be less efficient to train these models.Inference Speedups: Step Distillation, Consistency Models, and GANsSwyx [00:37:04]: Do you get the benefits of traditional diffusion speed-up? So for, images, there's LCM, LoRAs for, fine-tuning. There's, there's a lot of stuff that's beenEthan [00:37:15]: Flow matching.Swyx [00:37:16]: there's flow matching. There's a lot of stuff that's been done. there's some overlap that applies to diffusion on the inference side and stuff or?Ethan [00:37:23]: so the difference-- the inference side is a completely different story.Ethan [00:37:28]: I think for the training side, it might be a little bit hard to reduce that cost. And for the inference side, the biggest gain is from the distillation of these models. You can-- It's called step distillation, slightly different from knowledge distillation in LLMs. So you-- Typically, for flow matching models, you need like 100 steps or something. Like a distortion model even need even more, like 1,000 steps to generate a good image or video. A step distillation is try to learn to generate fewer step from the model itself. It's kind of like now we-- you use the full model to generate in 100 steps, and then you take a model that only generate 10 steps and let that model to learn from the perfect one.Ethan [00:38:25]: why this workSwyx [00:38:27]: Strong to weak seemingly.Ethan [00:38:28]: It is. It's kind ofSwyx [00:38:29]: DistillationEthan [00:38:29]: kind of like strong to weak. the-- from the modeling perspective, the strong model, the teacher model is trying to model the image and videos of inter-internet, and that distribution is extremely complex. But the step distilled model is just trying to learn from the teacher. The teacher is a model, and the size is fixed, as the distribution is much simpler than the whole internet. That's the intuition I have why step distillation can work. So usually these models serve in productions, they only run in a few steps. In Cosmos, I believe we have, we have like four step and eight steps. If you do some simpler task, image-image translation, it can even run in fewer step, like one step in Cosmos Transfer.Swyx [00:39:22]: I think this is the same intuition that guides a lot of the consistency model work. I sent you a link for, SCM. I don't know if you covered that. To me, that was actually one of, the most impressive papers I've ever seen from OpenAI.Swyx [00:39:34]: That this is the unifying grand concept of consistency models. I don't know if you have any comments on this.Ethan [00:39:41]: So there are, there are a few different approaches,Swyx [00:39:46]: Oh, yeah. Here it is.Swyx [00:39:47]: Two steps versus twenty or 100 steps, whatever. It's already done.Ethan [00:39:52]: So there are, there are a few different approaches, for example, consistency model, and there are also Actually, we shouldn't forget GAN. So GAN, actually, that was, that was the OG ofSwyx [00:40:05]: OGEthan [00:40:05]: step distillation ‘cause it trained just one step to begin with. So actually, a lot of, uh-- For example, there's a distribution matching distillation which use, which uses GAN, as one of the laws for distillation. It-- GAN just tells you, “Hey, generate an image,” and thenEthan [00:40:31]: it has a discriminator to tell, is this image real or not? So the model, the model just need to learn one of the distribution, not the full distribution. Because in training, the model is asked to reconstruct the ground truth image from the internet, which is extremely hard. And in-- When you're training GAN, it's a step process. It's just a, “Hey, you generate image. Does this image look as real as the image from the internet?” Which is a much simpler task. And, yeah, combining a lot of these approaches together, people typically do that, like consistency model and distribution matching and GAN, and we can get these few step models.Audio-Video Generation and Time AlignmentSwyx [00:41:21]: Then there's one step I wanted to add, which is audio and video.Ethan [00:41:26]: So, Grok Imagine zero point nine, I believe it's, it's a first audio video transmodel deployed at a large scale. SoSwyx [00:41:39]: And that was your first model?Ethan [00:41:40]: that was, Grok Imagine's first model. It's, it's audio video, joint generation. I think the hard part is, the modality alignment, ‘cause before this transmodel, we have, we have text to video alignment. We have this, correspondence between text and video. Typically, most of the VLMs, they understand images and videos. Video's very rare, and they don't understand audio mostly. And if you look at the audio generation on the LLM side, you can talk to them perfectly fine, but if you ask them to sing a song or something, it typically is not very good. Also, they don't have, they don't have music either. The hard part is thatUh, actually audio has two component. It has like a discrete component, a continuous component. The discrete component is like the language.Ethan [00:42:44]: So when we speak, it's just, someSwyx [00:42:47]: It's an ASR issue, yeah.Ethan [00:42:49]: It's, it's text token with some characteristics, I would say.Ethan [00:42:54]: But musicSwyx [00:42:56]: I think the speech guys would disagree with this.Swyx [00:42:57]: Like disfluencies and then,Vibhu [00:43:00]: There's tones you can get angry.Ethan [00:43:01]: Well, I say largely.Ethan [00:43:03]: the mu- but the music is completely different. It's, it's very continuous, and you cannot model them like discrete tokens in language models. this is like the hard part for models is, not to mention we have to align text, video, and audio together.Ethan [00:43:26]: SoVibhu [00:43:26]: How?Ethan [00:43:28]: So significant-- some significant challenges are like-- So first, like we talk about as the VLMs, they cannot understand most of them cannot understand audio.Ethan [00:43:39]: So you have to have some way to do the synthetic data generation for audio. You have to caption the model, and that involve, that involve synthetic data and human data effort a lot. And not just surprisingly, most of the LLMs are very bad at recognizing, like the beat, tone, and the details of the of music. They can, they can give some general prediction of which song is this, but it's very hard to describe the details of the music. like we mentioned in image generation, like you have to describe image as detailed as possible so that someone blind can reconstruct that. So here is like someoneVibhu [00:44:32]: DeafEthan [00:44:32]: someone deaf can reconstruct how the music sounds like without actually listening to it. Maybe you can think of it need to have the-- or they call the script.Vibhu [00:44:49]: Subtitles, yeah.Ethan [00:44:49]: You gotta have all the details of the music, and the dialogue.Vibhu [00:44:55]: So is the challenge there typically stuff like music and audio, or is it just Like is there a baseline? Okay, there's enough data where we can understand, narration, conversation, but there's nuances in audio that's where you hit all the data issues or is it just from stage zero, you just do it all right?Ethan [00:45:15]: So one important thing is like the alignment. So the model, the model has to know like the video and audio, the, uh-- it has to have a time-based alignment, like at which time step the video and the audio token correspond to each other. But we actually don't have this kind of alignment for most of the other modalities. If you think about like text and image, text and video, they are loosely aligned. So you can, you can have a description of what's going on in the video, but you don't have to exactly, You typically don't have exact description, oh, at, time step one second like what happened?Vibhu [00:46:02]: It's veryEthan [00:46:03]: At time step two second what happenedVibhu [00:46:03]: coarse. Yeah.Swyx [00:46:05]: So what was the ideal time step? You have to oblate it, and then it's like four seconds or something.Ethan [00:46:09]: So that comes down to how you design the model to, for the model to be aware of as a time, as a time modality. So the model is like a time aware. And that's something pretty unique if you think about LLMs. So if you ask LLM to complete a task, say they, uh-- you ask them and they will say, “Oh, this task will probably take twelve hours to complete,” and they come back in one hour. Say “I've already spent two days on this and I've exhausted everything.”Ethan [00:46:47]: So the LLMs them-themselves, they don't have a sense of time there.Vibhu [00:46:53]: I actually don't think that's just them not having a sense of time. I think it's somewhat based, right?Vibhu [00:46:58]: Like you tell someone, “Okay, go work on this feature. Go implement this,” there's a general understanding you would have of how long that would take without LLMs working at LLM speed, right? So you think back like two years ago, if I tell you to like build me like a new front end for latent space, have a search bar, have all this, you'll estimate that it'll take a few days, right?Vibhu [00:47:19]: So you tell an LLM, “Go build this.” It'll take me a few days. But I think it's somewhat grounded as opposed to them not having the best-- Not saying that they have a great understanding, but I think that example is like you can see where it comes from, right? You're trained on all over the text.Swyx [00:47:35]: They're, they're trying to estimate what a human would say.Vibhu [00:47:37]: because that's what the, that's what the data kind of represents. It's not themEthan [00:47:41]: It came from the corpus on the internet. People have a estimate of how much time.Vibhu [00:47:45]: And not even just in direct like training samples, right? Just your world understanding of tokens of how long stuff takes, right? Go read a book. It'll take you a while, right?Vibhu [00:47:56]: Even if you do nothing but read a book, it takes a few days. So yeah, LLM, I read it took me a few hours.Vibhu [00:48:01]: It'll take me a few hours to go through this research. But this is a tangent.Swyx [00:48:05]: Somewhat, yeah.Swyx [00:48:06]: This is a train of thought I haven't really expressed until now is, which is basically like a full world model must also be recursive, meaning that the participant in the world model must also be aware that they have a world model. which is like this whole recursive thing down the, down the line. but yes, and that the world model can be wrong and that they need to update it and blah. Yeah. We've, argued this on the, newsletter as well, that there needs to be sort of recursive or adversarial world models.World Models: Real-Time, Long-Horizon, Interactive VideoVibhu [00:48:34]: just, to ask, how do you define world model?Swyx [00:48:38]: Oh, yeah, let's go there.Ethan [00:48:40]: SoVibhu [00:48:40]: So just for context, we talked about, video generation, and then there's a-- if you say there's a distinction between world models, what's your, what's your definition? How do you see the two?Ethan [00:48:53]: So disclaimer, I'm not going to debate, what is world model. Yeah. there are many definitions, so I'll just talk about my definition. Since I came from the multi-model, multi-model domain, so mainly talking from video. So world model is like real-time interactive long horizon videos. So there are three parts. so we-- let's talk about them one by one. So the so interaction, so we just, we just look at Facebook and neural computer. So the interaction part of it, so you, world model can allow you to interact with them through keyboard, mouse, and maybe also voice. So these all is-- all is a modality. You can, you can interact with the model, and the model should respond reasonably. Second part is real time. So once you, once, say, you move your mouse, if, say, the world model generate a game, how fast can the game respond? So if you're like professional CS: GO players- -my say, oh, you have to respond- He's beginner within sub ten milliseconds or- Yeah even less. So that's not most of the- No, sixty FPS. Let's go. Oh, three hundred FPS. Oh, five hundred FPS. Wait. okay, yeah. I didn't do the math, but yeah, okay. Uh- Yeah, three hundred FPS, that's a three millisecond. So you have to respond- Oh, s**t. Okay. YeahEthan [00:50:29]: within a millisecond. Most of the video models cannot do that. Yeah. And, but if you, say, if you have a video model that is, say, like a digital human, the response time might be more generous. Maybe typically, for real-time voice interaction, it's like two hundred millisecond. So that's, that's much more generous. But even two hundred millisecond is pretty, it is pretty tricky, ‘cause remember we mentionedEthan [00:51:01]: you have this, temporal compression coming from the VAE. So if you, if you don't compress the temporal dimension, your sequence length is going to explode. So if you want to have this real-time, real-timeness in your model, you have to do is one context problem. And the third part is long horizon, ‘cause we-- if you're not going to just play with, video games just, a few seconds, most video models only a few seconds. We're going to play with minutes, hours. The model have to be able to generate long-form content.Ethan [00:51:42]: So putting these three together, it's, real-time, long horizon interactive videos. I think the final state will be, for example, like a video, a video version of Playbook, where you can, you can interact with, a neural computer. You move your mouse, and you click on the generative interface, and it will reply to you through pixels- generating in real time. But getting there, it's, it's a very long way to get there. So one of the first step, at Grok Imagine, where I led a small world model team there, was to build video extension. So, video extension- it's the first step of interactivity. Yeah. It's, it's the first step. Yeah. So it's the first step- You have it here, video editing, yeah. Yeah. Yeah. So the first step is because, this unlocks long horizon videos. Typically, for most of the video generation models, you give it a prompt or an image as an initial frame. You generate video, that's it. That's just, one time, done. And some creators would try to, use the last frame as a first frame for the second video. It can-- sometimes it works, but if you do it a few times, it says the quality would decrease. And- It doesn't have that context- Yeah over the full video, so the temporal- Yeah, exactly. Yeah, ‘cause you only gave it the last frame, of course, right? Yeah. Exactly. And- it's actually a pretty fun hack. if you've seen like- Oh, no, he's saying something better. Yeah. And for example, like Vue, I remember Vue 3 has like a second context of the last video. It is slightly better than using the last frame, but it has the same problem-- similar problem that it, the quality would decrease. if you extend a few times to, one minute, the video quality would look much worse than the first video. Second, another problem is that the model doesn't have long-range knowledge of, what's happening before. Say, if they generate some dialogue, some, two people speaking, and their voice might change, over some time, especially if the second conditioning, it does not cover the previous context. So these are the core challenges. So the Grok Imagine video extension, it has historical context of all of the previous generated videos. It can, It has, it has the context of, who is speaking and what objects have appeared and everything, having that to generate the next video. So if we naively do this, you can imagine, just, put all of the previous history video tokens into the context. The context lens will easily explode. Especially for video models, that can be like a few, a few million context, I would imagine- context lens. Yes.Yeah.Swyx [00:54:58]: Let's run with that.Ethan [00:54:59]: for example, like in Cosmos, I think just five seconds of video is like a fifty K or sixty K number of tokens. So like if you do, if you do fifty second, that's a five hundred K tokens. If you do longer than that, easily explode. This long horizon, problem was the first step we're trying to solve world model. It turns out people, yeah, people love video extension. Like a lot, a lot of the creators love using video extension to create longer form videos. This is the part I liked that you have a, you have an intermediate step toward the final goal instead of just a straight shot to the final version very much.Swyx [00:55:48]: But I can see you have a strong vision of where we want to end up.Long Context, Redundancy, and Efficient Interactive VideoVibhu [00:55:51]: Does it seem like it's an efficiency issue? okay, we're at a few million tokens context,. If you draw the parallel to language models, we had very short context, two thousand, eight thousand, then, you scale it up one million, ten million. sure, there's effective context, but at the end of the day, it's just what's it worth? sure, there's a whole training data side. In video, it might be slightly easier ‘cause we have a hundred million token video, right? Just take a movie with the full context there. Like is this efficiency from an inference standpoint that like it's expensive, but we know how to solve it? Or like why is this not the approach? So like my broader point was on your second point of world models, you say it needs to be interactive and live, right? You should be able to play a game and see the interaction live. So one thing I see with research is a lot of what you actually serve is different than what you build, right? So we talked about distillation. You train big model, you distill it, you do quantization, speculative decoding. We do all this stuff to serve it efficiently. Should we not just have a solution, like a world model that can interact well, do inference optimization, serve it, distill it secondary, so make it real time after you solve it? So like a-- another parallel is say, continual learning, right? What we need is someone to solve it and show it works inefficiently. Give it a few years, people will make it efficient. Same thing with regular attention, right? It worked. Over a few years, people have different forms of attention, and we've scaled it to be efficient at log context,? So kind of two things there, right? One is it seems like it works. You've scaled it. Can we not just scale it a lot more efficiently over time? Do we need a separate approach if this works? And same thing with interaction, right? if we can get it done, like if we can solve some way that it works, we can solve making it more efficient from an inference standpoint later.Ethan [00:57:53]: that's actually a very good point. So in videos, there's actually a lot of redundancies. So we solve a lot of the pixel redundancy from VE, but there's more redundancy in long range and long horizon videos. Say, if a character appear in the first clip and then it disappeared, it only reappear at the end of the video, you probably don't need the-- the context, like in the middle of the generation. So you only need that character, where you need. So that's why, I helped build another feature. It's a reference video.Vibhu [00:58:36]: Is it here?Swyx [00:58:36]: is it the same model release or different one?Ethan [00:58:39]: It's a different one.Ethan [00:58:41]: You probably need to search onSwyx [00:58:43]: I'll find itEthan [00:58:43]: X reference to video.Ethan [00:58:46]: So reference video allow you to like upload up to seven images as condition and generate the video. Say, if like I want-- it can, it can be characters or objects or even scenes. Say like I want, I want condition on, Sean's selfie and holding a bladeSwyx [00:59:07]: We have a dogEthan [00:59:08]: or whatever.Swyx [00:59:08]: We put the dog in the thing.Ethan [00:59:09]: you can put them there and the video models will generate the video from and copies the context over. So that can solve a lot of the problems there, like the long context problem. It doesn't need to have a very long context, but it's-- I feel like it's an intermediate solution. The modelSwyx [00:59:29]: It's cheating.Ethan [00:59:30]: the model should be able to like selectively know, where should I draw the references. So say if I want to generate a movie, I generate it autoregressive, like a ten second at a time or something. And now this character appear, I can look back to where it first appear and, bring that back. Yeah, this one, I put the references. Yeah, that's, Optimus, Einstein myself, Annie.Vibhu [01:00:02]: Oddly enough, I used Grok Search to find it, and it pulled your LinkedIn post. But yeah we found it.Ethan [01:00:08]: Interesting.Vibhu [01:00:10]: ButxAI's Underrated Work, Culture, and WatermarkingSwyx [01:00:11]: this is a problem. This is not your fault, but like XAI doesn't communicate all this work that you do very well because they just have the model release and then that's it. But actually, these details are very good.Swyx [01:00:22]: As far as I understand, everything you just described is state-art, like no one else has done it.Vibhu [01:00:30]: A lot of-- yeah, I have a lot moreSwyx [01:00:32]: And then, and then you just put this blog post with the cookies. I'm this is not enough,?Swyx [01:00:37]: but I, obviously this is like the high level numbers that people want to know. But no, okay, soVibhu [01:00:42]: And I wonder, like part of that is also some labs don't share research into what happens. And ifSwyx [01:00:50]: No, but this is literally bragging about how good they are, right?Swyx [01:00:54]: Like, why would you not say that you are capable of extending with full context? this is not a secret sauce. This is like we did the work. yeah, I don't know.Ethan [01:01:02]: different labs have slightly different communication styles.Swyx [01:01:07]: Anyway, if anyone from XAI is listening we are always happy to help you tell your story. Yeah, okay, so you did references, and I think, I think kind of the point you're, you're making is it is sort of like a kludge, right? this is-- you can do seven, but what about 100?Swyx [01:01:23]: Right? Then you need a completely different thing.Ethan [01:01:26]: So I think it's-- this is, a mechanism to, select the context from the history, and you might not put the entire history into the context. for example, there's a paper called Frame Pack, which haveEthan [01:01:41]: a heuristic that the latest history, the last one second, I put the entire history, and the history before that, I would, compress it and makes the video smaller. So they follow this pattern, this build overall pattern that the maximum sequence length is fixed. So the further you are from the current frame, you have a smaller image. So this is just a heuristic. I think it can be more automatic. The model is aware like which history part of it can be select. So this part of the research is actually being actively, worked on by a lot of people. It's also quite interesting. I feel this is actually, this part of long context is a little bit ahead of the LLM part.Ethan [01:02:31]: So for example, like in LLMs, if you-- so contexts keep growing. Let's say if you call tool and the tool call history is extremely long, that's still in context, and keep growing, keep growing. Even if you switch the topic to something else, the whole context was there. There are some agentic harnesses that help you to, say, prune the tool results and, prune Like when you, when you query a file, only show like the top 200 lines or something. Those were very heuristic-driven.Swyx [01:03:08]: For listeners, we did a write-up on the cloud code, leak where there are eight different kinds of pruning, including like you prune the tool results and all that. So you can, you can read up on that kind of thing.Ethan [01:03:17]: I think, one breakthrough in continual learning might be like a way to automatically, manage its own context.Swyx [01:03:27]: These are all heuristics, and they will be replaced by machine learning.Ethan [01:03:30]: InterestinglyVibhu [01:03:32]: TheEthan [01:03:32]: the same thing is being researched in both LLMs and video models.Vibhu [01:03:36]: The interesting thing is also like in the paper you showed, it's actually happening at the model level, right? Compared to like language models, sure, we have base attention, but we'll do our own compression, we'll do our own pruning, which is separate from model error.Vibhu [01:03:49]: Eventually, it all just boils in, hopefully.Swyx [01:03:52]: I think this is a form of like attention, but like also know sort of reasoning attention. I feel like that's different than normal attention.Swyx [01:04:03]: Does that, does that make sense?Ethan [01:04:04]: It's, it's different in the sense that attention, not to mention, set sparse attention aside,

The squad is complete again, and Sam arrives with a NeuroPod, cold plunge updates, red light therapy, Oura stats, and enough supplements to start a wellness startup. Then into the week's biggest tech stories: Google's new AI device and whether it's the Chromebook of the AI era or another doomed health-tech experiment, Meta's keystroke logging controversy, Microsoft's increasingly awkward OpenAI bet, why OpenAI and Anthropic are now sending engineers directly into enterprises to drive adoption, and what tools like OpenClaw, Py, and Codex actually do. Plus, Anthropic's eye-watering latest valuation, the clean girl aesthetic discourse, Brian Johnson chaos, and Sam personally buying Jackson Hole ski passes like it's 1997Chapters:00:46 Sam's NeuroPod, Oura Results & Biohacking Spiral03:33 Sam vs. Brian Johnson + The Female Biohacker Opportunity05:09 Oura Ring vs. Whoop + Google's Wearables Ambition07:00 Google's AI-First “Book” Laptop + DeepMind's Health Push10:30 Why Local AI Changes Everything (Speed, Cost & Compute)15:00 Where Is the OpenAI Consumer Device?16:00 Voice AI, Recording & the Future of Human-Computer Input20:30 Sam Built His Own Voice-to-AI App22:31 Meta's Keystroke Logging: Spy Games or Honeypot?24:00 Fake AI Jobs + Sam's “Fin Analytics” Prediction27:02 OpenAI & Anthropic's Enterprise Conversion Strategy29:31 The AI Backlash Is Real (Including UCF's Commencement Revolt)31:30 Microsoft's $100B OpenAI Problem39:31 Anthropic's Massive Raise + SF Real Estate Absurdity41:30 OpenClaw, Py & Codex: What Is a Harness?We're also on ↓X: https://twitter.com/moreorlesspodInstagram: https://instagram.com/moreorlessYoutube: https://youtu.be/-O3zyxR-wS0Connect with us here:1) Sam Lessin: https://x.com/lessin2) Dave Morin: https://x.com/davemorin3) Jessica Lessin: https://x.com/Jessicalessin4) Brit Morin: https://x.com/brit

durée : 02:01:11 - par : Jean-Baptiste Urbain - Trois ans après sa nomination à la tête du Châtelet, Olivier Py présente sa nouvelle saison qui fait la part belle à la comédie musicale, également à l'affiche actuellement avec "Top Hat". L'homme de théâtre n'a pas renoncé à mettre en scène : il propose bientôt un spectacle consacré à Nijinski. - réalisation : Yassine Bouzar, Max Dozolme, Julia Macarez, Morgane Tourreilles, Maxime Laporte, Valentin Lévy-Chaudet - invités : Olivier Py Dramaturge, metteur en scène, écrivain, acteur et réalisateur français, Pierre-Yves Lascar Directeur de label Vous aimez ce podcast ? Pour écouter tous les épisodes sans limite, rendez-vous sur Radio France

durée : 00:26:13 - par : Jean-Baptiste Urbain - Trois ans après sa nomination à la tête du Châtelet, Olivier Py présente sa nouvelle saison qui fait la part belle à la comédie musicale, également à l'affiche actuellement avec "Top Hat". L'homme de théâtre n'a pas renoncé à mettre en scène : il propose bientôt un spectacle consacré à Nijinski. - réalisation : Yassine Bouzar, Julia Macarez, Morgane Tourreilles, Maxime Laporte, Valentin Lévy-Chaudet - invités : Olivier Py Dramaturge, metteur en scène, écrivain, acteur et réalisateur français Vous aimez ce podcast ? Pour écouter tous les épisodes sans limite, rendez-vous sur Radio France

On this episode of the RIDINOUTALLDAY podcast.We talking about the worst piece of P***Y ever experienced. Yeah… we went there.The latest Tiger Woods crash got people talking. Reggie and Tiger Eskimo brothers? Kiki Shepard tribute.Bob Barker racist as sh*t.Pooh Shiesty… world's dumbest rapper? You decide.OnlyFans talk. Breaking down how the platform changed the game for creators.Conspiracy theory talk. Steven Spielberg ain't directing the landing. He said f*** that. That's why they not landing on the moon. He don't wanna be part of the conspiracy.And airline stewardess looking like WHO out here?All this and more on Episode 231 of the RIDINOUTALLDAY podcast.Available on all streaming platforms.Follow us on Instagram @mdavis @jones.boyzTap in at www.ridinoutallday.com.

!function(){try{var d=document.documentElement,c=d.classList;c.remove('light','dark');var e=localStorage.getItem('theme');if('system'===e||(!e&&true)){var t='(prefers-color-scheme: dark)',m=window.matchMedia(t);if(m.media!==t||m.matches){d.style.colorScheme = 'dark';c.add('dark')}else{d.style.colorScheme = 'light';c.add('light')}}else if(e){c.add(e|| '')}if(e==='light'||e==='dark')d.style.colorScheme=e}catch(e){}}()!function(){try{const t=localStorage.getItem("oai/apps/chatTheme/user-o92mEGCImh6FJHZE0l1Byw8L");document.documentElement.dataset.chatTheme="string"==typeof t?JSON.parse(t):void 0}catch(e){}}();Skip to contentJSChat historyNew chat⇧⌘OSearch chats⌘KImagesAppsDeep researchCodexProjectsYour chatsATP Weekly Indian Wells ReviewWTA Miami Open PreviewAlex Eala Indian Wells LossSEO Description for VideoHijikata Stuns BublikSEO Description for NorrieAlex Eala Indian Wells WinTsitsipas Career CrossroadsIndian Wells Tjen LossSpice tolerance comparisonWTA Weekly Indian Wells PreviewIndian Wells Preview 2026Indian Wells Draw ReactionBucsa Wins First WTA TitleLuciano Darderi ATP WinWTA Austin Final BreakdownATP Weekly Indian Wells PreviewTennis News and PreviewsTikTok Description TennisExample chat: Ask anythingJSJohn SilkFreeUpgradeGet PlusChatGPTGet PlusShareYou said:Podcast description for: ATP Weekly: Indian Wells review | Where are we at with the Sinner-Alcaraz rivalry? Is Medvedev back? Tien or Fonseca: How do they compare? How good was Draper-Djokovic?ChatGPT said:WritingIn this week's episode of ATP Weekly, we break down everything from Indian Wells and what it tells us about the current state of the men's tour.We start with a full Indian Wells review—the standout performances, surprises, and what the tournament revealed about the ATP landscape. Then we dive into the evolving Sinner–Alcaraz rivalry: where does it stand right now, and is it shaping up to define this era of men's tennis?We also discuss Daniil Medvedev's form and whether his recent level signals a true return to his best tennis. On the rising-stars front, we compare two of the most exciting young prospects on tour: Learner Tien and João Fonseca—how their games differ, their ceilings, and who might break through first.Finally, we analyze one of the most compelling matches of the tournament: Jack Draper vs. Novak Djokovic. How high was the level? What did it show about Draper's potential—and about Djokovic's current position in the field?All that and more on this week's episode of ATP Weekly.Is this conversation helpful so far?

Pyörien säilyttäminen sisällä herättää paljon tunteita, ja lopulta Esko ja Veronica ovat tästä täysin eri mieltä. Turussa on ollut mystinen viikko, ja tämä on Veronicalle erityisen sydäntä lähellä oleva aihe. Aiempi Voimariini-keskustelu herättää yhä tunteita, ja myös kuulijakokemuksia käydään edelleen läpi. Lisäksi kaksikko yrittää keksiä kasvisruoille vaihtoehtoisia nimiä esimerkiksi pihvin ja hedelmälihan tilalle. Lähdetäänkö vihville? Huumoria, viihdettä ja hulvattomia tarinoita! Esko Eerikäinen ja Veronica Verho kokoavat Radio Novan Iltapäivän parhaat palat yhteen Verho & Eerikäinen -podcastissa. Kuuntele myös livenä - Radio Novan Iltapäivä maanantaista torstaihin klo 14-18.

Suoraa puhetta johtaa Ruben Stiller. Keskustelemassa Taru Tujunen, Petja Kopperoinen ja Kaarina Hazard. Petja Kopperoinen pohtii ydinaseiden liikkeitä. Jos ydinrajoitteiden purkaminen toteutuu ja ydinaseita alkaa kulkea Suomen halki, millaisia ajatuksia se keskustelijoissa herättää? Ja nukkuvatko nämä yönsä paremmin tietäen, että kotimaassakin saattaa tulevaisuudessa olla ydinkärkiä? Taru Tujusen katse on tiukasti tulevaisuudessa. Jokainen sukupolvi vuorollaan puntaroi, mitä taitoja jatkossa pidetään arvossa. Entä tekoälyn ja humanoidirobottien maailmassa: Mitä sellaista ihminen osaa, jonka muuttuu arvokkaammaksi juuri siksi, että kone tekee kaiken muun? Kaarina Hazard nostaa pöytään julkisen keskustelun pelastamisen. Etenkin poliittisen somesisällön on katsottu algoritmien vuoksi olevan niin kallellaan, että sen katsotaan uhkaavan demokratiaa. Tähän päätyi Sitra. Samaan aikaan perinteiset viestimet näyttävät matkivan yhä enemmän somemaailmaa, eikä olla sen vastavoima. Hazard kysyykin, missä voimme jatkossa käydä julkista, yhteistä keskustelua? Oikaisu 11.3. Pyöreän pöydän lähetykseen: Lähetyksessä väitettiin, että viiteen Nato-maahan (Belgia, Italia, Saksa, Turkki ja Alankomaat), joihin ydinaseiden sijoittaminen olisi mahdollista, ei olisi niitä sijoitettu. Tämä ei pidä paikkaansa, kaikissa näissä maissa on sijoitettuna yhdysvaltalaisia ydinaseita.

Petr Pilát pochází z Pyšel u Benešova. Jeho tatínek ho na motorku posadil už ve třech letech. První závody jel o rok později. Když mu bylo třináct, objevil freestyle motokros. A protože byl od přírody talent, stal se záhy nejmladším jezdcem, který na motorce skočil backflip, tedy salto vzad. A další úspěchy na sebe nenechaly dlouho čekat. Potkalo ho i několik opravdu vážných zranění. Přesto ho nikdy ani nenapadlo, že by se freestyle motokrosu mohl vzdát.

Teppo Turkki on maamme johtava Aasian asiantuntija, jonka Itätuulen tuomisia -blogia seuraa pelkästään Suomessa sata tuhatta ihmistä. Kun Yhdysvallat on päivä päivältä suuremmassa kaaoksessa ja Eurooppa haukkoo henkeään, on selvää että Aasia vahvistuu. Pyöräytimme Tepon kanssa karttapalloa ja koetimme tarkastella maailmaa ja sen tilaa indopasifiselta alueelta käsin. Mitkä ovat silloin valtakulttuureita, mitkä vasallivaltioita? Miten vuosituhantinen perinne, keisarit, uskonnot, kollektivismi ja kolonialismi näkyvät tänä päivänä? Mitä Kiina haluaa? Entä Intia, entä Kaakkois-Aasian valtiot? Entä kuinka Aasian sisäinen dynamiikka toimii kokonaisuutena? Ja vielä, mikä on Venäjän rooli Kiinan ajattelussa? Ja lopuksi: mitä meidän Euroopassa ja Suomessa on syytä odottaa tapahtuvaksi ja miten me voimme siihen valmistautua? Paljon kysymyksiä – ja paljon vastauksia. Tervetuloa tähän ajatuksia niksauttelevaan keskusteluun maailman muutoksen mahdollisesti suurimmasta voimasta tällä hetkellä. Hyviä kuunteluhetkiä!  

In this episode of the ifa Show, general manager of Viridian Advisory Brett Arnol joins host Keith Ford to speak about how the firm approaches the Professional Year as a strategic talent pipeline rather than a compliance hurdle. Arnol explains that simply following the regulatory framework is not enough to turn a new adviser into a capable "advice‑ready" professional, and breaks down how Viridian maps each adviser's path over one, three and five years, using the PY as a foundation for long‑term career development, clarity of expectations, and strong retention. Tune in to hear: How listening, confidence, and the willingness to have difficult client conversations can be just as important as technical knowledge. Why clarity within the PY process is vital to setting advisers up for long-term success. How a well‑run PY program creates loyal, "homegrown" advisers, but firms must accept a three- to five-year horizon before candidates are fully productive.

This show has been flagged as Clean by the host. References in order of first mention Daytimer - https://www.daytimer.com/ PalmPilot - https://en.wikipedia.org/wiki/PalmPilot Gina Trapani - https://en.wikipedia.org/wiki/Gina_Trapani Todo landing page - http://todotxt.org/ Todo file format - https://github.com/todotxt/todo.txt Dropbox - https://www.dropbox.com/ Simpletask - https://github.com/mpcjanssen/simpletask-android/ QTodoTxt - https://github.com/QTodoTxt/QTodoTxt Synology DS220J NAS - https://global.download.synology.com/download/Document/Hardware/DataSheet/DiskStation/20-year/DS220j/enu/Synology_DS220j_Data_Sheet_enu.pdf Ice_recur - https://github.com/rlpowell/todo-text-stuff Py_recur - https://github.com/TASpinner/py_recur Microsoft todo - https://to-do.office.com/tasks/ Provide feedback on this episode.

Suoraa puhetta johtaa Ruben Stiller. Keskustelijoina Kaarina Hazard, Hilkka Olkinuora ja Pekka Seppänen. Täsmälleen samaan aikaan Pyöreä pöydän kanssa kokoontuivat Yhdysvaltain, Tanskan ja Grönlannin edustajat keskustelemaan siitä, kuka omistaa, kenet ja miksi. Hilkka Olkinuora haluaakin tämän kaiken kuohunnan keskellä kysyä ritarien ennustusta siitä, miten Trumpille lopulta käy? Viikonloppuna ei kukaan vähänkään mediaa seuraava suomalainen voinut välttyä uutisilta valtamerisoutaja Jari Saarion haaksirikosta Etelä-Atlantilla. Vaikka Saario pääsi turvaan, keskustelu jatkuu. Pekka Seppänen kysyy, mitä ajatuksia episodi herätti ja mitä opetuksia se tarjosi? Kaarina Hazard on huolissaan luku- ja kirjoitustaidon rapistumisen vaikutuksesta ihmisten ajattelukykyyn. Lyhytvideoiden ja tekoälyn aikakaudella ihmisten oman ajattelun sanotaan jäävän vähemmälle. Onko näin ja olemmeko tyhmistymisen tiellä?

Suoraa puhetta johtaa Maria Pettersson. Keskustelijoina Kaarina Hazard, Juha Itkonen ja Mika Pantzar. Kaarina Hazard haluaa kuulla ritarien ennustuksia siitä, mihin silmienvenyttelykuvista alkanut kohu lopulta johtaa. Jos tämä joulunalus jostain muistetaan, niin siitä, että maailma tuntee Suomen nyt valtiona, jossa toimitaan rasistisesti aasialaisia kohtaan. Pyöreän pöydän lähetystä seuraavana päivänä perussuomalaisten eduskuntaryhmä käsittelee asiaa sisäisesti. Rankaisevatko perussuomalaiset lopulta omiaan? Ja jos ei, niin saammeko hallituskriisin vielä ennen joulua? Juha Itkonen pohtii hyvinvointivaltion ympäristövaikutusta. Kansanedustaja Anna Kontula kirjoitti Helsingin Sanomissa, että hyvinvointivaltio olisi lähestymässä loppuaan ja törmäämässä ekologisiin rajoihin. Itkosen mielestä ylikulutus on kuitenkin vielä enemmän voimissaan esimerkiksi Yhdysvalloissa. Onko hyvinvointivaltio siis jotenkin erityisellä tavalla ekokriisiä kiihdyttävää ja pitäisikö sitä sen vuoksi alkaa purkamaan? Mika Pantzar pohtii saarnan asemaa nykypäivänä, vaikka kumpikaan Pyöreän pöydän pastorijäsenistä ei ole paikalla. Joulusaarnalla halutaan yleensä nostaa esiin yhteiskunnallinen ongelma tai muu asia, johon toivotaan muutosta. Ovatko nykyihmiset johdateltavissa toisenlaiseen käytökseen? Millaisen joulusaarnan kanssakeskustelijat haluaisivat kuulla tänä vuonna?

durée : 00:29:00 - Les Midis de Culture - par : Marie Labory - Pénétrer les portes de "La Cage aux folles" adaptée sur la scène du Châtelet par Olivier Py, c'est entrer dans un monde haut en paillettes, porté par Laurent Lafitte dans le personnage de Zaza, toute de plumes vêtue. - réalisation : Laurence Malonda - invités : Olivier Py Metteur en scène, Directeur du Théâtre du Châtelet

Rakennusala on montussa. Millä eväillä sieltä pääsee pois? Tänään korjaussarjansa esittää Suomen yrittäjät. Löytyykö vauhtia? Studiossa Yrittäjien Petri Malinen ja Skanskan Tuomas Särkilahti. Venejä vyöryttää sotilaitaan hyökkäyksessään teuraaksi tuhansien viikkovauhtia.Pyörittääkö Venäjä valloitusmyllyään aina, kun tilanne on otollinen? Studiossa sotahistorian dosentti Lasse Laaksonen ja sota-analyytikko Emil Kastehelmi. Markus Liimatainen juontaa, Petri Kejonen tuottaa, Seppo Toivonen ja Mari Sarolaihti toimittavat.

Puhetta johtaa Maria Pettersson. Keskustelemassa Maija Vilkkumaa, Kaarina Hazard ja Ruben Stiller. Kaarina Hazard tuo pöytään viikon uutisotsikoita hallinneen avustajajupakan. Sosiaali- ja terveysministeri Kaisa Juuso (ps) palkkasi vastikään eduskunta-avustajakseen oman poikansa ja kansanedustaja Jaana Strandman (ps) puolisonsa. Ilmiö ei ole uusi. Mutta kenet Pyöreän pöydän keskustelijat palkkaisivat? Miten he kehittäisivät avustajajärjestelmää niin, että homma pysyisi hallinnassa, työn sisältö ja vaatimukset määritelty ja kaikkien tiedossa? Ruber Stiller haluaa pöytäseurueen pureutuvan vihapuheeseen. Suomi sai asiassa pyyhkeitä, kun Euroopan rasismin ja suvaitsemattomuuden vastainen komissio (ECRI) arvosteli tuoreessa raportissaan vihapuheen kitkemistoimiamme riittämättömiksi. Vaikka toimia vihapuhetta vastaan meilläkin on tehty, onko niillä ollut sanottavaa vaikutusta? Onko sananvapauskäsitys muuttunut peruuttamattomasti? Maija Vilkkumaa kysyy, onko demokratia menossa pois muodista? Osa nuorempien sukupolvien edustajista ei enää usko demokraattiseen päätöksentekoon, kun maailmaa ravistuttavat kriisit vain jatkuvat. Kaivataan vahvaa johtajaa. Samaan aikaan esimerkiksi Kiina sysää ilmastotoimiaan eteenpäin, diktatuurissahan ei äänestäjien mielipiteitä tarvitse miettiä. Millaista demokratiaa 2.0 maailma nyt tarvitsee? Millä argumenteilla demoratiaa voisi tehokkaasti puolustaa?

Trailpodder jaksossa 137 pyörävelho Tom Nylund herättää vanhan Giant Talon -maastopyöräni puolessa tunnissa henkiin ja antaa vinkkejä pyörän kunnossapitoon. Alumiinirunko, ohjaustanko, satula, polkimet, 27 vaihdetta ja 27” renkaat ja vanteet pääsivät jatkoon puhdistuksen jälkeen. Pyörän pesu ja voitelu käytön jälkeen onkin tärkeää. Kaupungeissa saattaa olla pyörien omatoimisia huolto- ja pesupisteitä. Itse kävin pesemässä pyörän Helsingin Triplan pyöräparkissa (3 krs). Joustohaarukka edessä vaati ruosteen poistoa. Jarrulevyt ja takavaihtajan rissat puhdistettiin. Ketjut, takarataspakka ja jarrupalat piti uusia. Tällä pyörällä on vielä paljon ajonautintoa edessä!

Suomi ottaa käyttöön velkajarrun. Kaikki ymmärtävät, ettei velkaa voi kasvattaa loputtomasti – mutta miksi ja mihin velkajarrua oikeastaan tarvitaan? Kaarina Hazardin mielestä kyse on siitä, että jatkossa Suomen suunnasta eivät päätä kansalaiset, vaan virkamiehet. Miten muut ritarit suhtautuvat velkajarruun ja sen tarpeellisuuteen? Pyöreän pöydän nuorisoritarilainen Petja Kopperoinen haluaa tietää, mitä varttuneemmat kollegat ajattelevat eläkeiän nostamisesta 70 vuoteen – ei joskus tulevaisuudessa, vaan mieluiten heti. Ajatus ahdistaa yhtä, toinen sanoo antaa mennä vaan, ja kolmas näkee siinä jopa mahdollisuuksia. Mutta kuka ritarikunnasta on mitäkin mieltä? Pekka Seppänen nostaa pöydälle tuoreen aiesopimuksen, jonka mukaan "jokin yhdysvaltalainen taho kenties joskus tilaa suomalaisilta telakkayhtiöiltä joitakuita jäänmurtajia". Suomalaiset ovat revetä riemusta, mutta onko siihen oikeasti aihetta? Tiedätkö muuten mistä kolikosta löytyy jäänmurtajan kuva?

In this episode, host Sandy Vance sits with Derek Lo, CEO and Founder at Medallion, to explore how technology is reshaping one of the most overlooked but critical parts of healthcare: managing provider networks. Derek shares the story behind Medallion, why he set out to tackle the complexities of credentialing, licensing, and provider management, and how his team is using automation and AI to make life easier for both providers and organizations.Medallion builds software that simplifies the complexity of running a provider network. From credentialing to licensing, the platform helps organizations get providers seeing patients faster while offering greater efficiency, visibility, and control.In this episode, they talk about:How Medallion helps accurately manage provider networks with AI automationThe complicated and complex process of credentialing—and why it sparked Derek's journeyWhere Medallion comes into play in recruiting and hiring providersHow this benefits organizations beyond the administrative processThe future of AI management software in healthcareHow Medallion is helping drive transformation across the industryA Little About Derek:Derek Lo is the CEO and Founder of Medallion, the leading platform for provider network management. Since launching in 2020, he's grown the company to over 300 customers, built a 150+ person team, and raised $140 million from top investors like Sequoia Capital and Optum Ventures. A second-time founder, Derek previously built and sold Py to Hired.com in 2019. He's a Yale graduate in Computer Science and Statistics, a two-time Forbes 30 Under 30 honoree, and is driving Medallion's mission to simplify healthcare operations with AI-powered automation.

Pyöreän pöydän studiossa keskustelemassa Anu Koivunen, Mika Pantzar ja Juha Itkonen. Puhetta johtaa Maria Pettersson. Anu Koivunen nostaa keskusteluun Gazan tilanteen. Helsingin Sanomat uutisoi eilen (29.7.) gallup -kyselyn tuloksena, että hieman alle puolet (48 %) suomalaisista tunnustaisi Palestiinan. Vajaa neljännes vastaajista eli 24 prosenttia katsoi, että Suomen ei pitäisi tunnustaa Palestiinaa. Pitäisikö Suomen tunnustaa Palestiinan valtio? Miten te olisitte vastanneet kysymykseen ja mitä ajattelette ulkopoliittisen johdon (presidentti Alexander Stubbin) toukokuisesta toteamuksesta, että tunnustetaan sitten kun teolla on tehoa. Olemme kuulleet, että Ranska aikoo tunnustaa ja Iso-Britanniakin miettii. Mitä ajattelette meidän poliittisen johdon toiminnasta tässä asiassa? Mika Pantzar on valinnut aiheeksi Naton puolustusmenot. Nato-maat ovat sopineet puolustusmenojen nostamisesta 5 prosenttiin BKT:sta. Sopimus pitää sisällään 1.5 prosenttia kansallisesti harkittaviin investointeihin/hankkeisiin. Tämä tarkoittaa, että noin neljä miljardia euroa suunnataan "maanpuolustukseen" liittyviin menoihin. Mika pyytää ritareita ideoimaan kohteita, jotka sopisivat tähän löysään kategoriaan. Mitä te tekisitte neljällä "maanpuolustusmiljardilla"? Juha Itkonen ottaa puheeksi Jussi Pullisen Helsingin Sanomiin (27.7.) kirjoittaman kolumnin otsikolla "Journalismin tulee rikkoa sukupolvien siiloja". Pullisen mukaan Beatles, IShowSpeed (Yhdysvaltalainen striimaaja ja rap-artisti) ja sote mahtuvat kaikki uutismediaan, jonka tulisi olla eri sukupolvia yhdistävä tila. Pullinen piirtää tekstissään Suomesta kuvan poikkeuksellisen ikäsiiloutuneena yhteiskuntana. Hän viittaa kirjoituksessaan vuoden alussa julkaistuun E2 Tutkimuksen sukupolvibarometriin. Saatesanoissaan tutkimuksen tekijät kirjoittavat näin: "Suomalaisessa elämäntavassa eri sukupolvet ovat etääntyneet toisistaan. Palvelut, työelämä ja julkinen keskustelu eivät vahvista sukupolvien yhteistyötä, vaan oletuksena on pikemminkin, että ikä ominaisuutena määrittelee paitsi ihmisen tarpeita, myös esimerkiksi arvoja, voimavaroja, osallisuutta ja kiinnostuksen kohteita." Juha tiedustelee raadilta, koetteko te että näin on. Oliko ennen paremmin ja mitä asialle voisi tehdä?

Suoraa keskustelua luotsaa puheenjohtaja Maria Pettersson. Sananvapauttaan käyttämässä ovat Kaarina Hazard, Petja Kopperoinen ja Hilkka Olkinuora. Kaarina Hazard nosti pöydälle kuuman perunan: Miksi Palestiinan tunnustaminen on Suomelle niin vaikeaa? Kansa vaikuttaa olevan eduskuntaa myöten pääosin asian puolella. Tasavallan presidentti Alexander Stubb sanoi kuitenkin maanantaina Kultarannassa, ettei hän ”näe lisäarvoa” Palestiinan tunnustamiselle juuri nyt. Mitä Stubb tällä tarkoitti? Petja Kopperoinen on havainnut provider-kulttuurin nostavan päätään. Tällä tarkoitetaan siis sukupolvien takaista mallia, jossa perinteisesti miehen tehtävänä oli elättää ja naisen tehtävänä olla kaunis. Nyt näyttää siltä, että esimerkiksi Tiktokissa ovat nuoret miehet alkaneet liputtaa mallin puolesta. Mistä johtuu provider-kulttuurin uudelleennousu? Hilkka Olkinuora kysyy näin juhannusviikon kunniaksi kanssakeskustelijoilta, joko nämä ovat päässeet rantakuntoon. Millainen tämä kysymys ja tavoitteenasettelu on, sekä eettiseltä että esteettiseltä kannalta? Onko rantakuntoon pääseminen sukupuolittunut asia vai onko “velvoite” siihen kaikilla? On hyvä muistaa, että lihavuus itsessään on vaarallista. Miksi suhtaudumme siihen kuitenkin näin ulkonäkökeskeisesti? Pyöreä pöytä palaa kesätauolta keskiviikkona 30.7.

Suoraa puhetta johtaa Maria Pettersson. Keskustelijoina ovat Juha Itkonen, Hilkka Olkinuora ja Pekka Seppänen. Tarkennus ohjelman alussa olleeseen keskusteluun helsinkiläisestä kirjakaupasta: yliopiston kirjaston kiinteistön omistaa Helsingin yliopiston rahastot, joka ei liity Helsingin yliopiston ylioppilaskuntaan mitenkään. Juha Itkonen hämmästelee, mihin on sivistyksen arvostus hävinnyt. Aiheensa pohjustukseksi hän kertoo pari uutista viime viikoilta. Eduskunnan lukusalin kaunokirjallisuuskokoelma ollaan lakkauttamassa säästösyistä - vuositasolla säästöä saavutetaan noin 5000 euroa. Päätöksestä vastaava tieto- ja viestintäosaston johtaja Rainer Hindsberg perustelee ratkaisua sillä, että kaunokirjallisuuskokoelmalla on kansanedustajien keskuudessa kaikkien vähiten puolustajia.   Myös Kaisaniemessä Helsingin yliopiston pääkirjaston alakerrassa toimivan Rosebud-kirjakaupan tulevaisuus on epäselvä. Kirjakauppa haluaisi jatkaa tiloissa, mutta ilmeisesti vuokranantajalla Helsingin Yliopistokiinteistöt oy:llä on kiikarissa mahdollisesti vähän paremmin maksava vuokralainen. Nämä uutiset herättävat Juhan mielessä synkkiä ajatuksia. Suomessa sivistystä ei ole hänen elinaikana arvostettu yhtä vähän kuin nyt. Vähän joka suunnassa näkee puhdasta teknokratiaa ja excel-meininkiä. Pyöreän pöydän keskustelijoilta Juha kysyy, onko sivistyksen arvostus hävinnyt. Mitä mieltä olette? Hilkka Olkinuora haluaa puhua sähkölaudoista, joita ei pitäisi kutsua sähköpotkulaudoiksi, koska eihän niitä kukaan potkiskele. Parissa vuodessa on sähkölautaonettomuuksissa kuollut 4-5 ihmistä. Hilkka ihmettelee, miten monta henkilöä sähkölaudoilla on tänä kesänä kuoltava, ennen kuin sähkölaudat kielletään Suomessa. Mitä mielestänne pitäisi tehdä ja kenen tulisi tehdä? Entä miksi Suomi on näin nössö tässä asiassa? Pekka Seppänen kertoo Helsingin Sanomien julkaisseen tänään keskiviikkona 21.5.2025 ensimmäisen HS Hyviä uutisia -uutiskirjeen, joka sisältää vain hyviä uutisia. Pekka toteaa kaikkien Pyöreän pöydän ääressä tällä hetkellä istuvien jossakin vaiheessa elämäänsä syyllistyneen toimittajana työskentelemiseen. Pekka tiedusteleekin raatilaisilta, mitä mieltä he ovat hyvistä uutisista. Onko HS Hyvät uutiset -uutiskirjeen lanseeraus hyvä uutinen? Entä mitä tarkoittaa "hyvä uutinen"? Kuka sen määrittelee? Onko olemassa objektiivisesti katsottuna "hyviä uutisia"?

Ray and Dave battle it out with what wines best represent the DC Comics Characters Batman and Joker. Listen to find out which wine they each chose and who they thought chose the best bottle to represent that character.What wine would you have chosen for Batman and Joker?2019 Champagne Vauversin Rossingol Oger Grand Cru Blanc de Blancs (100% Chardonnay).2015 Lagier Meredith Mondeuse.2022 Domaine Jean Foillard "Côte du Py" Beaujolais Morgon (Gamay Noir).Ray's Homemade Cocktail using Del Maguey Mezcal Vida Single Village.

Suoraa puhetta johtaa Maria Pettersson. Keskustelijoina ovat Ruben Stiller, Mika Pantzar ja Maija Vilkkumaa. Ruben Stiller ihmettelee, miksi ihmeessä ihmiset ovat aivan spagettina Euroviisujen takia. Stiller tunnustaa, että on pitkäaikainen Euroviisujen boikotoija. Se on ainoa aate, joka hänellä on elämänsä aikana ollut. Erityisen tuohtunut hän on tapahtuman estetiikasta ja glitterin paljoudesta, kuvailee sen olevan eurotrashiä. Minkä takia Euroviisut on niin jättimäinen spektaakkeli? Entä mikä on tämän laulukilpailun funktio? Mika Pantzar haluaa puhua populismista. Mitä populismi oikeastaan tarkoittaa, sitä kritisoidaan paljon suomalaisessakin keskustelussa. Vasemmistolainen populismi on sitä, että epäillään rikkaiden ihmisten huijaavan ja vievän kaiken. Oikeistolainen populismi, varsinkin Yhdysvalloissa on ollut kulttuuritaisteluja ja tieteen sekä kaiken korkeakulttuurin kyseenalaistamista. Pitäisikö meidän Pyöreän pöydän jäsentenkin katsoa peiliin, edustammehan jonkinlaista eliittiä, kun saamme höpöttää radiossa. Pitäisikö katsoa peiliin, jotta näkisimme että ihmisillä on erittäin hyvä syy äänestää ns. populistisia liikkeitä? Suomessa on suuri määrä ihmisiä, jotka kokevat joutuneensa yhteiskunnan ulkopuolelle. Tarvitaanko Suomeen populistipuolue ja mitä ajattelemme ihmisistä, jotka heitä äänestävät? Maija Vilkkumaan aihe kumpuaa Oskari Onnisen Uusi juttu verkkomediaan kirjoittamasta artikkelista, joka käsitteli tekijänoikeuksia ja tekoälyä. Tällä hetkellä digijätit kouluttavat koko ajan tekoälyn kielimalleja sellaisilla teksteillä, jotka ovat jo olemassa. Internetissä on olemassa jo valtavia varjokirjastoja, joita ainakaan eurooppalaisen tekijänoikeuslainsäädännön mukaan ei saisi käyttää ilmaiseksi ja kysymättä lupaa. Mitä mieltä olette, onko oikein vai väärin, että digijätit näin toimivat ja mitä kaikkea tästä voi seurata?

John Fardy was joined by Michelle Lawlor, Deirdre Molumby and Russell Alford with thanks to Marks and Spencer!This week's booze:Vesevo Beneventano Falanghina, ItalyDomaine Trenel Morgon Cote du Py

durée : 00:05:15 - Comme personne - Marie Emmanuelle Py, militaire sur le porte-avions nucléaire Charles de Gaulle, est devenue coach en entreprise. Sa spécialité : philosophie existentielle et quête de sens. Elle intervient dans de grandes entreprises, mais aussi bénévolement auprès de militaires pour leur retour à la vie civile.

Suoraa puhetta johtaa Maria Pettersson. Keskustelijoina ovat Hilkka Olkinuora, Ruben Stiller ja Mika Pantzar. Hilkka Olkinuora muistelee, että Pyöreän pöydän ympärilläkin on toisinaan joku parahtanut "vaalit ovat ihmisen parasta aikaa". Toisaalta tänään alkoi alue- ja kuntavaalien ennakkoäänestys, johon oli vaikea saada ehdokkaita ja ilmeisesti myös äänestäjiäkin. Hilkka kysyy ytimekkäästi, miksi kannattaa äänestää. Ruben Stiller ryydittää puhettaan lyömällä nyrkkiä pöytään ja samalla kertoo innostuneesti, kuinka presidentti Stubb oli seitsemän tuntia Yhdysvaltain presidentti Trumpin seurassa. Suomi puttasi itsensä maailmankartalle! Kotimaahan palattuaan Stubb ilmoitti, että meidän pitää valmistautua avaamaan uudelleen poliittiset suhteet Venäjään ja sieltä on tullut tähän ajatukseen innostunut vastaus. Ruben kysyy, missä geopoliittisessa tilanteessa Suomi on nyt. Mitä mieltä olette golf-diplomatian tuloksesta? Entä kuinka otettuja olette presidenttimme taitavuudesta? Mika Pantzar nurisee hyväntahtoisesti Pyöreän pöydän puheenaiheiden Trump keskeisyydestä, hänestä emme tunnu pääsevän eroon. Mikakin tarttuu aiheeseen, siirtämällä Trumpin ajattelua vähän Suomeen. Trumpilla on slogan "järjen ja totuuden palauttaminen Amerikan historiaan". Mikan kysymys kuuluu, mitä ajattelette jos Suomeen tulisi ääriliikkeen presidentti. Minkälainen tulisi olemaan tällaisen ajan historiankirjoitus? Mitä tehtäisiin uudelle Kansallismuseolle, mitä sinne tuotaisiin ja mitä poistettaisiin? Onko meillä kyky tuottaa uutta historinkirjoitusta ja kuka sitä tuottaa? Kertokaa millaisia ideoita teillä on uudelle ääriliikkeen edustajalle? Kenen patsaita poistetaan ja kenelle niitä pystytetään?

durée : 00:07:53 - Nouvelles têtes - par : Mathilde Serrell - Il enflamme le Châtelet depuis quelques jours dans une version déjantée du chef d'oeuvre d'Ibsen porté par la musique de Grieg, le comédien Bertrand de Roffignac est l'invité de Mathilde Serrell.

Dave Harding and Mike Schmidt are joined by Bastien Teinturier and Joost Jager discuss ⁠Newsletter #342⁠.News● Allowing mobile wallets to settle channels without extra UTXOs (0:59) ● Continued discussion about an LN quality of service flag (13:14) Changes to services and client software● Ark Wallet SDK released (40:28) ● Zaprite adds BTCPay Server support (40:57) ● Iris Wallet desktop released (41:21) ● Sparrow 2.1.0 released (41:41) ● Scure-btc-signer 1.6.0 released (42:38) ● Py-bitcoinkernel alpha (43:48) ● Rust-bitcoinkernel library (44:30) ● BIP32 cbip32 library (45:56) ● Lightning Loop moves to MuSig2 (46:22) Notable code and documentation changes● Bitcoin Core #27432 (47:01) ● Bitcoin Core #30529 (48:29) ● Bitcoin Core #31384 (49:42) ● Core Lightning #8059 (50:52) ● Core Lightning #7985 (53:41) ● Core Lightning #7887 (54:32) ● Eclair #2967 (26:06) ● Eclair #2979 (32:24) ● Eclair #3002 (34:32) ● LDK #3575 (57:35) ● LDK #3562 (23:51) ● BOLTs #1205 (26:13)

VOV1 - Hôm nay (29/1) tức mùng 1 Tết Nguyên đán Ất Tỵ, tại xã Pờ Y, huyện Ngọc Hồi, Bộ Chỉ huy Bộ đội Biên phòng tỉnh Kon Tum tổ chức Lễ chào cờ chủ quyền quốc gia cột mốc biên giới Việt Nam - Lào - Campuchia.

Varda Etienne nous révèle pourquoi elle est à boutte de sa cinquantaine ! Phil Roy nous explique pourquoi l’aviation et la musique son indissociable pour lui. Marie-Soleil Michon nous fait un segment : J’ai une question pour PY. BONNE ÉCOUTE !

In this episode, Manny and I talk about what is a "Trophy"?  in this case, we keep the conversation about trophy whitetails. We start off discussing what the various scoring clubs consider a trophy score is, such as B&C, P&Y, and SCI. We all know that a trophy can maean many things to many different hunters, and I think we can all agree that it is a matter of perspective. We talk about how comercialized hunting has changed the game and in turn has changed the hunting stories of today.  For me yes, score may set the perameters that classify what a tropy is, but I believe it is the story behind the harvest that is the true trophy. Contact us at sotxoutdoors@gmail.com or via FB and IG @sotxoutdoors 

Accountable Care Organizations (ACOs) have made great strides in enhancing patient care and reducing costs, saving $3.1 billion in 2023. The recently released report based on ACO performance in 2023 (PY2023) provides valuable insights into where ACOs are excelling. In this episode of Value-Based Care Insights, Dan Marino sits down with Sarah Kachur PharmD, MBA, BCACP, Executive Director of Strategy and Solutions at Johns Hopkins Population Health Analytics, to discuss the key findings of the report and what they reveal about the future of ACOs. Gain insights into the priorities of the most successful ACOs (in PY 2023) and their predictions for changes and challenges in 2025 and beyond. LinkedIn https://www.linkedin.com/in/sarahgracekachur/ Skachur2@jh.edu Hopkins ACGs: https://www.hopkinsacg.org/ Illustra Health from Johns Hopkins: https://illustra.health/

Ep 117 - Sarah Kachur- ACOs Saved Billions in 2023: Can They Sustain This Beyond 2025? Accountable Care Organizations (ACOs) have made great strides in enhancing patient care and reducing costs, saving $3.1 billion in 2023. The recently released report based on ACO performance in 2023 (PY2023) provides valuable insights into where ACOs are excelling. On this episode Dan sits down with Sarah Kachur PharmD, MBA, BCACP, Executive Director of Strategy and Solutions at Johns Hopkins Population Health Analytics, to discuss the key findings of the report and what they reveal about the future of ACOs. Gain insights into the priorities of the most successful ACOs (in PY 2023) and their predictions for changes and challenges in 2025 and beyond. To stream our Station live 24/7 visit www.HealthcareNOWRadio.com or ask your Smart Device to “….Play Healthcare NOW Radio”. Find all of our network podcasts on your favorite podcast platforms and be sure to subscribe and like us. Learn more at www.healthcarenowradio.com/listen

durée : 00:39:22 - Les Midis de Culture - par : Marie Sorbier - Après plus d'une cinquantaine de mises en scène à l'opéra, Olivier Py revient à ses racines pour reprendre ses mises en scène acclamées du "Dialogue des Carmélites" de Poulenc au Théâtre des Champs Élysées et "The Rake's Progress" de Stravinsky à l'Opéra Garnier. - réalisation : Laurence Malonda - invités : Olivier Py Metteur en scène, Directeur du Théâtre du Châtelet

O Pyšnou princeznu je obrovský zájem v zahraničí. Novou českou animovanou pohádku uvidí diváci po celém světě. „Nedokážu si představit, že bychom dnes dělali znova hraný film, takže animovaná verze se nabízela sama. Chtěli jsme to přiblížit dětem,“ popisuje jeden z režisérů animované Pyšné princezny David Lisý. Jak vznikala animovaná verze slavného hraného filmu Bořivoje Zemana z 50. let?Všechny díly podcastu Host Radiožurnálu můžete pohodlně poslouchat v mobilní aplikaci mujRozhlas pro Android a iOS nebo na webu mujRozhlas.cz.

O Pyšnou princeznu je obrovský zájem v zahraničí. Novou českou animovanou pohádku uvidí diváci po celém světě. „Nedokážu si představit, že bychom dnes dělali znova hraný film, takže animovaná verze se nabízela sama. Chtěli jsme to přiblížit dětem,“ popisuje jeden z režisérů animované Pyšné princezny David Lisý. Jak vznikala animovaná verze slavného hraného filmu Bořivoje Zemana z 50. let?

  Discover what's new for the ACA Market in PY 2025. HealthSherpa's Ross Baker joins Sarah to discuss changes and updates for the upcoming Open Enrollment Period.   Mentioned in this episode: HealthSherpa Official Site: https://www.healthsherpa.com/agents/features   Follow HealthSherpa on Social: Facebook: https://www.facebook.com/healthsherpas/ Instagram: https://www.instagram.com/healthsherpas/ LinkedIn: https://www.linkedin.com/company/healthsherpa/ X: https://x.com/healthsherpas YouTube: https://www.youtube.com/@Healthsherpas    Additional Ritter Resources: How Much Can Agents Make Selling Under-65 Insurance: https://link.chtbl.com/ASG620 Simplify Marketplace Enrollments with HealthSherpa: https://ritterim.com/blog/simplify-marketplace-enrollments-with-healthsherpa/ Steps to Get Ready for OEP: Federal & State Exchanges: https://link.chtbl.com/ASG6215 The Complete Guide to Selling Affordable Care Act Insurance Plans: https://ritterim.com/aca-ebook/   Follow Us on Social! Ritter on Facebook, https://www.facebook.com/RitterIM Instagram, https://www.instagram.com/ritter.insurance.marketing/ LinkedIn, https://www.linkedin.com/company/ritter-insurance-marketing TikTok, https://www.tiktok.com/@ritterim X, https://twitter.com/RitterIM and Youtube, https://www.youtube.com/user/RitterInsurance      Sarah on LinkedIn, https://www.linkedin.com/in/sjrueppel/ Instagram, https://www.instagram.com/thesarahjrueppel/ and Threads, https://www.threads.net/@thesarahjrueppel  Tina on LinkedIn, https://www.linkedin.com/in/tina-lamoreux-6384b7199/   Contact the Agent Survival Guide Podcast! Email us ASGPodcast@Ritterim.com or call 1-717-562-7211 and leave a voicemail.

TEMATY ODCINKA

Lee La, PY and Dave from The Damn Truth joined Jay and Sharon in the CHOM 97 7 studio to preview their newest song, Love Outta Luck.

【主播】贤鱼，Storm，于渤，Seven，卓然（@贤鱼不会写段子）（@Storm徐风暴）（@累死于渤算了）（@Seven不是个英文名）（@单口喜剧卓然）咦！你这人咋就认死理嘞！这期聊的【轴】，固执，死板，钻牛角尖，死守教条，不懂变通，它形象地描绘了一种人在原地打转的状态。他们固执地坚持着自己的看法和做法，践行传统的逻辑和思维方式，即使偶尔会吃些苦头，即使外界已经发生了翻天覆地的变化，他们也依然我行我素，不为所动。但也是这种轴的特质，让轴的人在生活中展现出了一种独特的魅力。他们不畏惧世俗的眼光和评判，勇敢地活出自我。他们平静如水，懂得获取满足内心的快乐。今天啊，咱们聊聊【轴】，还就跟您较较这个真儿！【shownotes】8:47宁愿在上海打拼也不回家继承农场！噢！你的生活给我过吧！20:08生活中没苦硬吃的人，你就轴吧！看看到时候苦涩哽咽了谁的喉！33:50滞留吉隆坡机场三天两夜，现实版人生中转站！37:05生活中的没福硬享！倔的要死！大众说你是中心轴！59:55我的个人坚持！以及轴的好处！周PY！你不许再biaji嘴了！1:15:45亲密关系中的底线，爱情中的坚守，这方面，我轴上江楼！1:27:40我不轴了，我不较劲了，算了，其实也挺好~无论如何转动，我们都能找到平衡~【演出推荐】【购票：喜剧联盒国vx小程序】贤鱼【上海】【10.1周二19:30上海喜联道馆】【单口喜剧专场】【我打不赢爱情】狒狒【上海】【10.4周五19:30上海喜联道馆】【单口喜剧主打秀】【拉拉家常先行版】小橘【上海】【10.6周日19:30上海喜联道馆】【单口喜剧主打秀】【互动方式】微信公众号，微博，抖音，小红书，B站：@嘻谈录【参加录制】微信公众号：嘻谈录【进听友群】和主播聊天，获取最新消息，领取节目周边，添加嘻谈录小助手微信：xitanlu2【背景音乐】来自YouTube音乐平台Dan Lebowitz -Ready and Steady Vigilante - Veneno【制作】：Wizzy（@WizzyWizzyWizzy）【视频剪辑】：陈嘉炜【社群运营】：马野【公众号】：鱼宝鱼【视觉设计】：Mila【商务合作】请加vx：xitanlu2

In this profoundly moving episode, Mick Hunt explores Patrick Young's incredible journey of resilience and transformation. From basketball courts to life's unexpected challenges, Patrick shares his experiences and insights on overcoming adversity, finding new purpose, and helping others through his foundation. Patrick Young's Background: Former SEC Defensive Player of the Year, professional athlete in the NBA, and Euroleague, now a broadcaster and author. Defining Moments: Patrick reflects on the life-changing car accident and his profound realization about identity and purpose beyond physical abilities. Discussion Topics:Patrick's transition from professional sports to overcoming personal adversities, including his paralyzing car accident.His ongoing recovery journey, the launch of his book "Sit to Rise," and the foundation of the PY 4 Foundation aiding those facing life-altering events.Discuss identity, resilience, and his motivational efforts to inspire others facing challenges.Key Quotes:"The pit that you're in doesn't stop your purpose.""Life is full of choices, and despite challenges, we have the power to choose our path."Next Steps:Learn More: Visit the PY 4 Foundation website to discover how you can support those facing life-altering challenges.Reflect: Consider your resilience and the choices that define your path.Engage: Share how Patrick's story inspires you to face your own challenges with courage using #MickUnplugged.Connect & Discover:Instagram: instagram.com/PatricYoung4Facebook: facebook.com/py4foundationWebsite: PlayersforGood.comWebsite: PatricYoung.comX: PatricYoung4

Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: (Approximately) Deterministic Natural Latents, published by johnswentworth on July 21, 2024 on LessWrong. Background: Natural Latents: The Math, Natural Latents: The Concepts, Why Care About Natural Latents?, the prototypical semantics use-case. This post does not assume that you've read all of those, or even any of them. Suppose I roll a biased die 1000 times, and then roll the same biased die another 1000 times. Then... Mediation: The first 1000 rolls are approximately independent of the second 1000 given the bias (to reasonable precision). Redundancy: I can estimate the die's bias (to reasonable precision) with high confidence from either the first or second 1000 rolls. The die's bias is therefore a natural latent, which means it has various nice properties. Minimality: The bias is the smallest summary of all the information about the first 1000 rolls relevant to the second 1000 (and vice-versa). Maximality: The bias is the largest piece of information which can be calculated from the first 1000 rolls and also can separately be calculated from the second 1000 rolls. Any other variable which satisfies the above properties must tell us (approximately) the same information about the die rolls as the bias. Furthermore, the bias is a(n approximate) deterministic natural latent: the die's bias (to reasonable precision) is approximately determined by[1] the first 1000 die rolls, and also approximately determined by the second 1000 die rolls. That implies one more nice property: Uniqueness: The bias is the unique-up-to(-approximate)-isomorphism latent which has the above properties, making it a natural Schelling point for communication between agents. We've proven all that before, mostly in Natural Latents: The Math (including the addendum added six months after the rest of the post). But it turns out that the math is a lot shorter and simpler, and easily yields better bounds, if we're willing to assume (approximate) determinism up-front. That does lose us some theoretical tools (notably the resampling construction), but it gives a cleaner foundation for our expected typical use cases (like e.g. semantics). The goal of this post is to walk through that math. Background Tool: Determinism in Diagrams We're going to use diagrammatic proofs, specifically using Bayes nets. But it's non-obvious how to express (approximate) determinism using Bayes nets, or what rules diagrams follow when determinism is involved, so we'll walk through that first. This diagram says that Y is (approximately) determined by X: Intuitively, the literal interpretation of the diagram is: X mediates between Y and Y, i.e. Y itself tells me nothing more about Y once I already know X. That only makes sense if X tells me everything there is to know about Y, i.e. Y is determined by X. In the approximate case, we express the approximation error of the diagram as a KL-divergence, same as usual: ϵDKL(P[X=x,Y=y,Y=y']||P[X=x]P[Y=y|X=x]P[Y=y'|X=x]) If you get confused later about what it means to have two copies of the same variable in a diagram, go back to that line; that's the definition of the approximation error of the diagram. (One way to view that definition: there's actually two variables Y and Y', but P says that Y and Y' always have the same value.) That approximation error simplifies: DKL(P[X=x,Y=y,Y=y']||P[X=x]P[Y=y|X=x]P[Y=y'|X=x]) =DKL(P[X=x,Y=y]I[y=y']||P[X=x]P[Y=y|X=x]P[Y=y'|X=x]) =x,y,y'P[X=x,Y=y]I[y=y'](log(P[X=x,Y=y]I[y=y'])log(P[X=x]P[Y=y|X=x]P[Y=y'|X=x])) =x,yP[X=x,Y=y](log(P[X=x,Y=y])log(P[X=x]P[Y=y|X=x]P[Y=y|X=x])) =x,yP[X=x,Y=y]log(P[Y=y|X=x]) =H(Y|X) So the diagram says Y is determined by X, and the approximation error of the diagram is the entropy H of Y given X - i.e. the number of bits required on average to specify Y once one already knows X. Very intuitive! The Dangly Bit Lemma Intuitiv...

We're at the garage in the city living large this week as we eat pizza and cookies then discuss The Babysitter. We agree that some of the filmmaking choices detract from the film's likability, but it's still a good watch. With laughs all the way through, this film is fun, entertaining, and uncomplicated. Watch it. Don't be a P*$$Y.

本期播客为「情感电台」的下半期，这期只有一位听众连线，然而他的故事却相当炸裂，各种出轨劈腿狗血剧情层出不穷，内容相当毁三观，建议大家从「听八卦」角度收听。*郑重声明：听众陈述内容非本台观点，本台强烈谴责一切渣男行为！/ 本期主播：马里奥、沙拉包、小宋/ 剪辑/封面/文案：沙拉包/ 本期内容： 上大学时我「脚踏四条船」 出轨被发现后从男女朋友变成PY 一千天纪念日被发现出轨 被认真谈的女友发好人卡后彻底开摆 我每天都要聊10个女孩子 确定关系那天她告诉我她是TS 我又认识了不少CD 我被前女友的现男友骂了 社交软件的利与弊

(Deuxième et dernier épisode) Le mercredi 25 juin 2014, Marc Isoird se rend au commissariat de Sète dans l'Hérault pour signaler la disparition de son petit frère, Patrick qui n'a plus donné signe de vie depuis deux jours. Le 23 juin, Patrick, 49 ans, avait un rendez-vous devant le cimetière du Py, à Sète, avec Audrey Louvet, une ancienne amante qu'il avait rencontrée plusieurs années auparavant. Entendue au commissariat, cette femme de 32 ans raconte avoir été déposée au rendez-vous par un ami qui s'appelle Rémi Chesne. Or, après vérification, les policiers découvrent qu'il existe un lien entre les deux hommes. Quelques années plus tôt, en 2009, Patrick Isoird avait été entendu par la police dans le cadre d'une enquête sur la mort par pendaison d'une certaine Nadège Chesne, la femme de Rémi Chesne. Patrick Isoird travaillait avec elle à l'hôpital de Sète et deux jours avant sa mort, ils s'étaient rapprochés lors d'une soirée.Dans Crime story, la journaliste Clawdia Prolongeau raconte cette enquête avec Damien Delseny, chef du service police-justice du Parisien.Écoutez Crime story sur toutes les plates-formes audio : Apple Podcast (iPhone, iPad), Google Podcast (Android), Amazon Music, Podcast Addict ou Castbox, Deezer, Spotify.Crédits. Direction de la rédaction : Pierre Chausse - Rédacteur en chef : Jules Lavie - Ecriture et voix : Clawdia Prolongeau et Damien Delseny - Production : Thibault Lambert, Barbara Gouy et Raphaël Pueyo - Réalisation et mixage : Julien Montcouquiol - Musiques : Audio Network - Archives : INA. Documentation. Cet épisode de Crime story a été préparé en puisant dans les archives du Parisien, avec l'aide de nos documentalistes. Nous avons aussi exploité les articles de Midi Libre. Hébergé par Acast. Visitez acast.com/privacy pour plus d'informations.

(Premier épisode) Le mercredi 25 juin 2014, Marc Isoird se rend au commissariat de Sète dans l'Hérault pour signaler la disparition de son petit frère, Patrick qui n'a plus donné signe de vie depuis deux jours. Le 23 juin, Patrick, 49 ans, avait un rendez-vous devant le cimetière du Py, à Sète, avec Audrey Louvet, une ancienne amante qu'il avait rencontrée plusieurs années auparavant. Entendue au commissariat, cette femme de 32 ans raconte avoir été déposée au rendez-vous par un ami qui s'appelle Rémi Chesne. Or, après vérification, les policiers découvrent qu'il existe un lien entre les deux hommes. Quelques années plus tôt, en 2009, Patrick Isoird avait été entendu par la police dans le cadre d'une enquête sur la mort par pendaison d'une certaine Nadège Chesne, la femme de Rémi Chesne. Patrick Isoird travaillait avec elle à l'hôpital de Sète et deux jours avant sa mort, ils s'étaient rapprochés lors d'une soirée.Dans Crime story, la journaliste Clawdia Prolongeau raconte cette enquête avec Damien Delseny, chef du service police-justice du Parisien.Écoutez Crime story sur toutes les plates-formes audio : Apple Podcast (iPhone, iPad), Google Podcast (Android), Amazon Music, Podcast Addict ou Castbox, Deezer, Spotify.Crédits. Direction de la rédaction : Pierre Chausse - Rédacteur en chef : Jules Lavie - Ecriture et voix : Clawdia Prolongeau et Damien Delseny - Production : Thibault Lambert, Barbara Gouy et Raphaël Pueyo - Réalisation et mixage : Julien Montcouquiol - Musiques : Audio Network - Archives : INA. Documentation. Cet épisode de Crime story a été préparé en puisant dans les archives du Parisien, avec l'aide de nos documentalistes. Nous avons aussi exploité les articles de Midi Libre. Hébergé par Acast. Visitez acast.com/privacy pour plus d'informations.

Why is Python so popular? There is plenty of room for debate on this but one solid reason is it's easy to adopt, easy to use, and caters to people who are not quite developers/data scientists but need to do some computing. Do you know where there largest untapped set of that group hang out? Excel. That's why it's super exciting that Python is now going to be built directly into Excel. Just go into a cell and type =PY and you're off writing full Python 3 code that is backed by a lite Anaconda distribution of Python. And we have Dr. Sarah Kaiser here to give us the rundown on Python in Excel. Episode sponsors Posit Pybites PDM Talk Python Courses Links from the show Sarah's website: sckaiser.com Sarah on Mastodon: @crazy4pi314@mathstodon.xyz Get started with Python in Excel: microsoft.com Python in SQL Server: microsoft.com 8 of the Biggest Excel Mistakes of All Time: blog.hurree.co Security and Python in Excel: microsoft.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to us on YouTube: youtube.com Follow Talk Python on Mastodon: talkpython Follow Michael on Mastodon: mkennedy