Podcasts about retort

An exciting week for Tom, who tells Nate about his company Hortus AI, whose mission is to make AI accountable to local communities. We cover a lot of classic Retort themes as Tom makes a case for what's missing from AI development, and how models could be more healthily integrated into everyday people's lives.Press release: https://hortus.ai/wp-content/uploads/2025/02/Hortus-AI-Press-Release_Feb25.pdfCompany website: https://hortus.ai/Link to whitepaper: https://www.newamerica.org/rethinkai/policy-papers/a-sustainable-path-for-ai-development/

We're back! Tom and Nate catch up after the Thanksgiving holiday. Our main question was -- what were the biggest AI stories of the year? We touch on the core themes of the show: infrastructure, AI realities, and and antitrust. The power buildout to scale out AI is going to have very real long-term impacts.Some links this week:* Ben Thompson's, The End of the Beginning: https://stratechery.com/2020/the-end-of-the-beginning/* Miles Brundage's Substack: https://milesbrundage.substack.com/p/why-im-leaving-openai-and-what-im* Stochastic Parrots paper: https://dl.acm.org/doi/10.1145/3442188.3445922Thanks for listening! Get The Retort (https://retortai.com/)…… on YouTube: https://www.youtube.com/@TheRetortAIPodcast… on Spotify: https://open.spotify.com/show/0FDjH8ujv7p8ELZGkBvrfv?si=fa17a4d408f245ee… on Apple Podcasts: https://podcasts.apple.com/us/podcast/the-retort-ai-podcast/id1706223190… Follow Interconnects: https://www.interconnects.ai/… email us: mail@retortai.com

An all Enterprise round of debates! Was humanity ready to explore the stars or were the Vulcans right? Did the Xindi get off too easy for their attack on Earth? Is Phlox the best doctor character in Trek? These and many more! Live Long and Podcast presents Locutors of Trek: Debate 9; Season 1, Episode 16 Date of Podcast: November 10, 2024 THIS EPISODE'S PODCASTERS Davan Skelhorn, Dave Puxley, Dave Mader, Kevin Millard, Jody Simpson, Adam Woodward A PROUD MEMBER OF THE UNITED FEDERATION OF PODCASTS PRODUCER Davan Skelhorn Check us out online at https://www.ufpodcasts.com/livelongan... Streaming live on Twitch, Youtube and Facebook: Twitch Channel: / livelongandpodcast YouTube Channel: / livelongandpodcast Facebook Page: / pg LiveLongAndPodcast Audio version available wherever you get your audio podcasts. Listen to Spotify: https://open.spotify.com/show/0yIEMJh... Listen via Anchor: https://anchor.fm/livelongandpodcast #StarTrek #LocutorsOfTrek #LiveLongAndPodcast #UFP #UnitedFederationofPodcasts

Live Long and Podcast presents Locutors of Trek: Debate 9; Season 1, Episode 16 Date of Podcast: November 10, 2024 THIS EPISODE'S PODCASTERS Davan Skelhorn, Dave Puxley, Dave Mader, Kevin Millard, Jody Simpson, Adam Woodward A PROUD MEMBER OF THE UNITED FEDERATION OF PODCASTS PRODUCER Davan Skelhorn Check us out online at https://www.ufpodcasts.com/livelongan... Streaming live on Twitch, Youtube and Facebook: Twitch Channel: / livelongandpodcast YouTube Channel: / livelongandpodcast Facebook Page: / pg LiveLongAndPodcast Audio version available wherever you get your audio podcasts. Listen to Spotify: https://open.spotify.com/show/0yIEMJh... Listen via Anchor: https://anchor.fm/livelongandpodcast #StarTrek #LocutorsOfTrek #LiveLongAndPodcast #UFP #UnitedFederationofPodcasts

Tom and Nate catch up on the happenings in AI. Of course, we're focused on the biggest awards available to us as esteemed scientists (or something close enough) -- the Nobel Prizes! What does it mean in the trajectory of AI for Hinton and Hassabis to carry added scientific weight. Honestly, feels like a sinking ship. Some links:* Schmidhuber tweet: https://x.com/SchmidhuberAI/status/1844022724328394780* Hinton "I'm proud my student fired Sam": https://x.com/Grady_Booch/status/184414542282424329000:00 Introduction04:43 Criticism of AI-related Nobel Prize awards09:06 Geoffrey Hinton's comments on winning the Nobel Prize18:14 Debate on who should be credited for current AI advancements25:53 Changes in the nature of scientific research and recognition34:44 Changes in AI safety culture and company dynamics37:27 Discussion on AI scaling and its impact on the industry42:21 Reflection on the ongoing AI hype cycleRetort on YouTube: https://www.youtube.com/@TheRetortAIPodcastRetort on Twitter: https://x.com/retortaiRetort website: https://retortai.com/Retort email: mail at retortai dot com

The Humanimals get back to the mission. It's time to make some plans. Dungeons and Dracon Beams is an Animorph's Dungeons and Dragons 5th Edition adventure with an entirely different team of Animorph's in an alternate homebrewed world.  What would happen if the events of the Yeerk invasion played out differently? The players will shape their own story, in their own city, fighting their own war. Will they make the change? Join Savannah, Kamren, Dylan, Zac, and Aximili as they gain the power to turn into any animal they can touch and go claw to tail blade with Visser Three and the Yeerk Empire. New chapters of D&DB will stream every other Sunday at twitch.tv/seezydrop with VODs and a podcast released the following week on Youtube. Join us and catch all the action and drama live! If you would like to support us and get extra content and a bunch of other perks, you can join our The Dungeons and Dracon Beams Patreon! The 5$ tier gets you exclusive episodes!   Cast Dylan - Jenna - @Jennachil Savannah - Alex - @AlexandBirds Zac - Nate - @SplintersmithNC GM - Austin - @DNDBPOD   Music “Beer Can” by Alisia from Pixabay “City Lights (Chill Lo-Fi)”  by Oleksii Holubiev from Pixabay “Anticipation”  by Florentina Popper from Pixabay “Summer Fun Energetic Pop Upbeat”  by Dana Music from Pixabay “Take Me to a New Land”  by Geoff Harvey from Pixabay “Pop Punk - Take Me Away” by Tech Oasis from Pixabay “Chilling at the beach.” Captain Phillips from Pixabay “Thinking Overture” by DSTechnician from Pixabay “Reflective Serious Piano (Rain On The Station)” by Ashot Danielyan from Pixabay “Saka”  by Jerome Chauvel from Pixabay “Terror y Suspenso - Autor: Marcos Molina” by Marcos Molina from Pixabay “Strange Unsettling Dream - Instrumental” by free-to-use-audio from Pixabay Sound Effects “ambiance, field-recording, market, street, people, crowd, city, walking_2” by FALL-E of freesound.org https://freesound.org/people/FALL-E/sounds/695643/ “College Ambiance” by Benjamin152 of freesound.org https://freesound.org/people/Benjamin152/sounds/445114/ “Car_window_open.wav” by AndrewAlexander of freesound.org https://freesound.org/people/AndrewAlexander/sounds/369057/ “Sink Wet Sounds.wav” by RutgerMuller of freesound.org https://freesound.org/people/RutgerMuller/sounds/50755/ “Bass Drop - Nasty 001” by GnoteSoundz of freesound.org https://freesound.org/people/GnoteSoundz/sounds/169713/ “Boiling liquid-Retort in action” by arnaud coutancier of freesound.org https://freesound.org/people/arnaud%20coutancier/  

In this episode of the Celebrate Kids podcast, Dr. Kathy delves into the challenges and disappointments that can come with parenting. She addresses the difficult emotions parents may face, such as doubt, loss of freedom, and shattered dreams. Dr. Kathy explores how to find hope, courage, and resolve in facing these challenges while honoring the Lord. The episode also touches on a viral video exposing a subreddit called Regretful Parents, prompting discussions about parental regrets and struggles. Tune in for an insightful discussion on navigating the darker aspects of parenting with faith and perseverance.

Welcome to another episode of the Multifamily Collective! Today, we're exploring a topic that's rapidly reshaping the multifamily industry: AI Readiness. Inspired by a recent CreTech episode, we'll explore how AI is not just a future concept but a present reality, ready to transform operations in real time. The accelerating pace of AI development is mind-blowing. One panelist suggested a scenario where you could voice a prompt to an AI tool to analyze utility expenses and receive actionable advice. This capability isn't in the distant future—it's here now with technologies like Lori from MoveOn.AI. Imagine asset managers having the power to optimize operations instantly with voice commands. Our industry is on the brink of a revolution, yet many aren't aware of the tools already available. The challenge isn't just about developing new tech but staying updated with what's out there. Even if you had a dedicated team, the rapid advancements in AI would still be hard to keep up with. In this fast-evolving landscape, staying connected to the right networks and influencers is crucial. Those who can adapt and integrate these innovations will lead the way. If you found this discussion insightful, like, subscribe, and hit the notification bell so you never miss an update. Stay ahead in the multifamily game with us! To stay abreast of the latest, visit https://multifamilycollective.com #Multifamily #AI #PropTech #RealEstate #AssetManagement #Innovation #TechTrends #Leadership #FutureOfWork #mikebrewer #multifamilycollective #multifamilymentoring #multifamilycoaching #multifamilypodcast #leadership #OpenAi #multifamilymedianetwork --- Support this podcast: https://podcasters.spotify.com/pod/show/mike-brewer/support

Don't skip parts one and two!   Part one: Youtube: https://youtu.be/z-oEql7mcxs?si=g7IiuNPhF9KFPImb Audio only: https://www.dotheology.com/e/cn-pt-1   Part two: Youtube: https://youtu.be/Ad8zdlT-png?si=UwIltzsvkGVZVO8b Audio only: https://www.dotheology.com/e/cn-part-2   In this third episode of a multi-part series, Jeremy dives into the commission given to the Church within the Christian Nationalism worldview. Though it can be difficult to find Christian Nationalists going on-record about what they believe the Church has been tasked to do in the here-and-now, Jeremy shares the sound bites he's found and analyzes what he hears.   Subscribe to the podcast: https://linktr.ee/DoTheology Check out Foundations Media: https://foundationsmedia.org   https://dotheology.com https://store.dotheology.com https://www.buymeacoffee.com/DoTheology   Contact Us: show@dotheology.com https://twitter.com/dotheology https://facebook.com/dotheology   0:00 Music 1:13 Introduction 4:53 Analyzing Definitions: What & How 6:40 Statement on Christian Nationalism 8:53 Stephen Wolfe 12:00 Doug Wilson 15:07 Joel Webbon 18:20 A.D. Robles 21:02 Voddie Baucham and Tom Ascol 24:05 GotQuestions.org 26:11 Jeff Durbin 28:43 CN Makes Political Reform the Church's Goal 31:10 Jeff Durbin Again 35:47 Andrew Sandlin 42:04 CN Believes We Convert Nations as Nations 46:39 Jeff Durbin Once More 49:21 Dusty Deevers 52:09 CN Desires to Legislate the Law of Moses 55:05 Joel Webbon Again 59:38 Mayberry & Norman Rockwell's America 1:02:53 CN Are Prolific Content Creators 1:05:43 The Moscow Billboard Campaign 1:07:04 The Failure of ChristIsLord.com 1:10:35 Retort 1: We're Not Tasked with World Domination 1:19:08 Retort 2: Public Reform Is Secondary to Evangelism 1:25:55 Conclusion

Does the GR do journalism or commentary?

Subscribe on Patreon and hear this week's full patron-exclusive episode here: https://www.patreon.com/posts/103298391 Beatrice, Jules and Artie discuss the findings, impact, and fallout of “The Cass Review,” the recently released 388 page NHS-commissioned report that has led media figures across the political spectrum to proclaim that the evidence supporting transition for young people is flawed. But the report, and the ensuing debate, are asking all the wrong questions. Get Health Communism here: www.versobooks.com/books/4081-health-communism Find Jules' new book here: https://www.versobooks.com/products/3054-a-short-history-of-trans-misogyny Runtime 1:42:06, 29 April 2024

All the girls are finally back in studio and some would call that a real treat. Jake & Griff try not to rub in their trip to ireland in the faces of Mart & Stots. They discuss deeply if wearing a hat indoors is acceptable. If your into retorts today is a good day for you. This episode is brought to you by GOODR. Sunglasses for your face. Also LONG DRINK. DRINKS FOR YOUR MOUTH. Lastly this episode is brought to you by Manscaped. Scape your man.

#DallasCowboys Fish For Breakfast 4/04: 30 Visits News, Ryan Leaf's Retort, 'The Dallas Chiefs'?! ✭ SUBSCRIBE to the NEW Fish Report Podcast here: https://www.dspmediaonline.com/show/the-dallas-cowboys-fish-report/ ✭ STRAIGHT DOPE. NO BULLSH. ✭ ✭ UNCLE FISH STORE https://shorturl.at/gJPS2 ✭ FISH SPORTS GEAR www.fishsportsnetwork.com Listen on the Go, 24/7! Download the NEW Fan Stream Sports APP on iOS and Android! Follow FISH on X: @FishSports #DallasCowboysReport CowboysSI.com https://www.si.com/nfl/cowboys/

#DallasCowboys Fish For Breakfast 4/04: 30 Visits News, Ryan Leaf's Retort, 'The Dallas Chiefs'?! ✭ SUBSCRIBE to the NEW Fish Report Podcast here: https://www.dspmediaonline.com/show/the-dallas-cowboys-fish-report/ ✭ STRAIGHT DOPE. NO BULLSH. ✭ ✭ UNCLE FISH STORE https://shorturl.at/gJPS2 ✭ FISH SPORTS GEAR www.fishsportsnetwork.com Listen on the Go, 24/7! Download the NEW Fan Stream Sports APP on iOS and Android! Follow FISH on X: @FishSports #DallasCowboysReport CowboysSI.com https://www.si.com/nfl/cowboys/

Steve says the judicial supremacists on the right need to shut up. Then, Megan Basham from the Daily Wire joins the show to talk about how leftist ideologies are infiltrating the church through a new "Bible study" developed by David French and others. In Hour Two, Theology Thursday studies the second session of "Know Thy Enemy: A Nefarious Bible Study." Finally, Ana Hibbs joins the show to give the guys another three non-political questions. Learn more about your ad choices. Visit megaphone.fm/adchoices

In 2023 we did a few Fundamentals episodes covering Benchmarks 101, Datasets 101, FlashAttention, and Transformers Math, and it turns out those were some of your evergreen favorites! So we are experimenting with more educational/survey content in the mix alongside our regular founder and event coverage. Pls request more!We have a new calendar for events; join to be notified of upcoming things in 2024!Today we visit the shoggoth mask factory: how do transformer models go from trawling a deeply learned latent space for next-token prediction to a helpful, honest, harmless chat assistant? Our guest “lecturer” today is ; you might know him from his prolific online writing on and Twitter, or from his previous work leading RLHF at HuggingFace and now at the Allen Institute for AI (AI2) which recently released the open source GPT3.5-class Tulu 2 model which was trained with DPO. He's widely considered one of the most knowledgeable people on RLHF and RLAIF. He recently gave an “RLHF 201” lecture at Stanford, so we invited him on the show to re-record it for everyone to enjoy! You can find the full slides here, which you can use as reference through this episode. Full video with synced slidesFor audio-only listeners, this episode comes with slide presentation along our discussion. You can find it on our YouTube (like, subscribe, tell a friend, et al).Theoretical foundations of RLHFThe foundation and assumptions that go into RLHF go back all the way to Aristotle (and you can find guidance for further research in the slide below) but there are two key concepts that will be helpful in thinking through this topic and LLMs in general:* Von Neumann–Morgenstern utility theorem: you can dive into the math here, but the TLDR is that when humans make decision there's usually a “maximum utility” function that measures what the best decision would be; the fact that this function exists, makes it possible for RLHF to model human preferences and decision making.* Bradley-Terry model: given two items A and B from a population, you can model the probability that A will be preferred to B (or vice-versa). In our world, A and B are usually two outputs from an LLM (or at the lowest level, the next token). It turns out that from this minimal set of assumptions, you can build up the mathematical foundations supporting the modern RLHF paradigm!The RLHF loopOne important point Nathan makes is that "for many tasks we want to solve, evaluation of outcomes is easier than producing the correct behavior". For example, it might be difficult for you to write a poem, but it's really easy to say if you like or dislike a poem someone else wrote. Going back to the Bradley-Terry Model we mentioned, the core idea behind RLHF is that when given two outputs from a model, you will be able to say which of the two you prefer, and we'll then re-encode that preference into the model.An important point that Nathan mentions is that when you use these preferences to change model behavior "it doesn't mean that the model believes these things. It's just trained to prioritize these things". When you have preference for a model to not return instructions on how to write a computer virus for example, you're not erasing the weights that have that knowledge, but you're simply making it hard for that information to surface by prioritizing answers that don't return it. We'll talk more about this in our future Fine Tuning 101 episode as we break down how information is stored in models and how fine-tuning affects it.At a high level, the loop looks something like this:For many RLHF use cases today, we can assume the model we're training is already instruction-tuned for chat or whatever behavior the model is looking to achieve. In the "Reward Model & Other Infrastructure" we have multiple pieces:Reward + Preference ModelThe reward model is trying to signal to the model how much it should change its behavior based on the human preference, subject to a KL constraint. The preference model itself scores the pairwise preferences from the same prompt (worked better than scalar rewards).One way to think about it is that the reward model tells the model how big of a change this new preference should make in the behavior in absolute terms, while the preference model calculates how big of a difference there is between the two outputs in relative terms. A lot of this derives from John Schulman's work on PPO:We recommend watching him talk about it in the video above, and also Nathan's pseudocode distillation of the process:Feedback InterfacesUnlike the "thumbs up/down" buttons in ChatGPT, data annotation from labelers is much more thorough and has many axis of judgement. At a simple level, the LLM generates two outputs, A and B, for a given human conversation. It then asks the labeler to use a Likert scale to score which one it preferred, and by how much:Through the labeling process, there are many other ways to judge a generation:We then use all of this data to train a model from the preference pairs we have. We start from the base instruction-tuned model, and then run training in which the loss of our gradient descent is the difference between the good and the bad prompt.Constitutional AI (RLAIF, model-as-judge)As these models have gotten more sophisticated, people started asking the question of whether or not humans are actually a better judge of harmfulness, bias, etc, especially at the current price of data labeling. Anthropic's work on the "Constitutional AI" paper is using models to judge models. This is part of a broader "RLAIF" space: Reinforcement Learning from AI Feedback.By using a "constitution" that the model has to follow, you are able to generate fine-tuning data for a new model that will be RLHF'd on this constitution principles. The RLHF model will then be able to judge outputs of models to make sure that they follow its principles:Emerging ResearchRLHF is still a nascent field, and there are a lot of different research directions teams are taking; some of the newest and most promising / hyped ones:* Rejection sampling / Best of N Sampling: the core idea here is that rather than just scoring pairwise generations, you are generating a lot more outputs (= more inference cost), score them all with your reward model and then pick the top N results. LLaMA2 used this approach, amongst many others.* Process reward models: in Chain of Thought generation, scoring each step in the chain and treating it like its own state rather than just scoring the full output. This is most effective in fields like math that inherently require step-by-step reasoning.* Direct Preference Optimization (DPO): We covered DPO in our NeurIPS Best Papers recap, and Nathan has a whole blog post on this; DPO isn't technically RLHF as it doesn't have the RL part, but it's the “GPU Poor” version of it. Mistral-Instruct was a DPO model, as do Intel's Neural Chat and StableLM Zephyr. Expect to see a lot more variants in 2024 given how “easy” this was.* Superalignment: OpenAI launched research on weak-to-strong generalization which we briefly discuss at the 1hr mark.Note: Nathan also followed up this post with RLHF resources from his and peers' work:Show Notes* Full RLHF Slides* Interconnects* Retort (podcast)* von Neumann-Morgenstern utility theorem* Bradley-Terry model (pairwise preferences model)* Constitutional AI* Tamer (2008 paper by Bradley Knox and Peter Stone)* Paul Christiano et al. RLHF paper* InstructGPT* Eureka by Jim Fan* ByteDance / OpenAI lawsuit* AlpacaEval* MTBench* TruthfulQA (evaluation tool)* Self-Instruct Paper* Open Assistant* Louis Castricato* Nazneen Rajani* Tulu (DPO model from the Allen Institute)Timestamps* [00:00:00] Introductions and background on the lecture origins* [00:05:17] History of RL and its applications* [00:10:09] Intellectual history of RLHF* [00:13:47] RLHF for decision-making and pre-deep RL vs deep RL* [00:20:19] Initial papers and intuitions around RLHF* [00:27:57] The three phases of RLHF* [00:31:09] Overfitting issues* [00:34:47] How preferences get defined* [00:40:35] Ballpark on LLaMA2 costs* [00:42:50] Synthetic data for training* [00:47:25] Technical deep dive in the RLHF process* [00:54:34] Projection / best event sampling* [00:57:49] Constitutional AI* [01:04:13] DPO* [01:08:54] What's the Allen Institute for AI?* [01:13:43] Benchmarks and models comparisonsTranscriptAlessio [00:00:00]: Hey everyone, welcome to the Latent Space podcast. This is Alessio, partner and CTO in Residence at Decibel Partners, and I'm joined by my co-host Swyx, founder of Smol AI.Swyx [00:00:15]: Hey, and today we have Dr. Nathan Lambert in the house. Welcome.Nathan [00:00:18]: Thanks guys.Swyx [00:00:19]: You didn't have to come too far. You got your PhD in Berkeley, and it seems like you've lived there most of the time in recent years. You worked on robotics and model-based reinforcement learning on your PhD, and you also interned at FAIR and DeepMind. You bootstrapped the RLHF team at Hugging Face, and you recently joined the Allen Institute as a research scientist. So that's your quick bio. What should people know about you that maybe is not super obvious about you on New LinkedIn?Nathan [00:00:43]: I stay sane in various insane sport and ultra-endurance sport activities that I do.Swyx [00:00:50]: What's an ultra-endurance sport activity?Nathan [00:00:52]: Long-distance trail running or gravel biking. Try to unplug sometimes, although it's harder these days. Yeah.Swyx [00:00:59]: Well, you know, just the Bay Area is just really good for that stuff, right?Nathan [00:01:02]: Oh, yeah. You can't beat it. I have a trailhead like 1.2 miles from my house, which is pretty unmatchable in any other urban area.Swyx [00:01:11]: Pretty excellent. You also have an incredible blog, Interconnects, which I'm a fan of. And I also just recently discovered that you have a new podcast, Retort.Nathan [00:01:20]: Yeah, we do. I've been writing for a while, and I feel like I've finally started to write things that are understandable and fun. After a few years lost in the wilderness, if you ask some of my friends that I made read the earlier blogs, they're like, oh, this is yikes, but it's coming along. And the podcast is with my friend Tom, and we just kind of like riff on what's actually happening on AI and not really do news recaps, but just what it all means and have a more critical perspective on the things that really are kind of funny, but still very serious happening in the world of machine learning.Swyx [00:01:52]: Yeah. Awesome. So let's talk about your work. What would you highlight as your greatest hits so far on Interconnects, at least?Nathan [00:01:59]: So the ones that are most popular are timely and or opinion pieces. So the first real breakout piece was when April and I also just wrote down the thing that everyone in AI was feeling, which is we're all feeling stressed, that we're going to get scooped, and that we're overworked, which is behind the curtain, what it feels to work in AI. And then a similar one, which we might touch on later in this, was about my recent job search, which wasn't the first time I wrote a job search post. People always love that stuff. It's so open. I mean, it's easy for me to do in a way that it's very on-brand, and it's very helpful. I understand that until you've done it, it's hard to share this information. And then the other popular ones are various model training techniques or fine tuning. There's an early one on RLHF, which is, this stuff is all just like when I figure it out in my brain. So I wrote an article that's like how RLHF actually works, which is just the intuitions that I had put together in the summer about RLHF, and that was pretty well. And then I opportunistically wrote about QSTAR, which I hate that you have to do it, but it is pretty funny. From a literature perspective, I'm like, open AI publishes on work that is very related to mathematical reasoning. So it's like, oh, you just poke a little around what they've already published, and it seems pretty reasonable. But we don't know. They probably just got like a moderate bump on one of their benchmarks, and then everyone lost their minds. It doesn't really matter.Swyx [00:03:15]: You're like, this is why Sam Altman was fired. I don't know. Anyway, we're here to talk about RLHF 101. You did a presentation, and I think you expressed some desire to rerecord it. And that's why I reached out on Twitter saying, like, why not rerecord it with us, and then we can ask questions and talk about it. Yeah, sounds good.Nathan [00:03:30]: I try to do it every six or 12 months is my estimated cadence, just to refine the ways that I say things. And people will see that we don't know that much more, but we have a bit of better way of saying what we don't know.Swyx [00:03:43]: Awesome. We can dive right in. I don't know if there's any other topics that we want to lay out as groundwork.Alessio [00:03:48]: No, you have some awesome slides. So for people listening on podcast only, we're going to have the slides on our show notes, and then we're going to have a YouTube version where we run through everything together.Nathan [00:03:59]: Sounds good. Yeah. I think to start skipping a lot of the, like, what is a language model stuff, everyone knows that at this point. I think the quote from the Llama 2 paper is a great kind of tidbit on RLHF becoming like a real deal. There was some uncertainty earlier in the year about whether or not RLHF was really going to be important. I think it was not that surprising that it is. I mean, with recent models still using it, the signs were there, but the Llama 2 paper essentially reads like a bunch of NLP researchers that were skeptical and surprised. So the quote from the paper was, meanwhile, reinforcement learning known for its instability seemed a somewhat shadowy field for those in the NLP research community. However, reinforcement learning proved highly effective, particularly given its cost and time effectiveness. So you don't really know exactly what the costs and time that Meta is looking at, because they have a huge team and a pretty good amount of money here to release these Llama models. This is just the kind of thing that we're seeing now. I think any major company that wasn't doing RLHF is now realizing they have to have a team around this. At the same time, we don't have a lot of that in the open and research communities at the same scale. I think seeing that converge would be great, but it's still very early days. And the other thing on the slide is some of Anthropic's work, but everyone knows Anthropic is kind of the masters of this, and they have some of their own techniques that we're going to talk about later on, but that's kind of where we start.Alessio [00:05:17]: Can we do just a one-second RL version? So you come from a robotics background, which RL used to be, or maybe still is, state-of-the-art. And then now you're seeing a lot of LLM plus RL, so you have the gym fans, Eureka, you have MPU, which we had on the podcast when they started with RL. Now they're doing RL plus LLMs. Yeah. Any thoughts there on how we got here? Maybe how the pendulum will keep swinging?Nathan [00:05:46]: I really think RL is about a framing of viewing the world through trial and error learning and feedback, and really just one that's focused on thinking about decision-making and inputs in the world and how inputs have reactions. And in that, a lot of people come from a lot of different backgrounds, whether it's physics, electrical engineering, mechanical engineering. There are obviously computer scientists, but compared to other fields of CS, I do think it's a much more diverse background of people. My background was in electrical engineering and doing robotics and things like that. It really just changes the worldview. I think that reinforcement learning as it was back then, so to say, is really different. You're looking at these toy problems and the numbers are totally different, and everyone went kind of zero to one at scaling these things up, but people like Jim Phan and other people that were... You saw this transition in the decision transformer and papers and when people are trying to use transformers to do decision-making for things like offline RL, and I think that was kind of like the early days. But then once language models were so proven, it's like everyone is using this tool for their research. I think in the long run, it will still settle out, or RL will still be a field that people work on just because of these kind of fundamental things that I talked about. It's just viewing the whole problem formulation different than predicting text, and so there needs to be that separation. And the view of RL in language models is pretty contrived already, so it's not like we're doing real RL. I think the last slide that I have here is a way to make RLHF more like what people would think of with RL, so actually running things over time, but a weird lineage of tools that happen to get us to where we are, so that's why the name takes up so much space, but it could have gone a lot of different ways. Cool.Alessio [00:07:29]: We made it one slide before going on a tangent.Nathan [00:07:31]: Yeah, I mean, it's kind of related. This is a...Swyx [00:07:35]: Yeah, so we have a history of RL.Nathan [00:07:37]: Yeah, so to give the context, this paper really started because I have this more diverse background than some computer scientists, such as trying to understand what the difference of a cost function or a reward function and a preference function would be without going into all of the details. Costs are normally things that control theorists would work with in these kind of closed domains, and then reinforcement learning has always worked with rewards that's central to the formulation that we'll see, and then the idea was like, okay, we now are at preferences, and each step along the way there's kind of different assumptions that you're making. We'll get into these, and those assumptions are built on other fields of work. So that's what this slide is going to say, it's like RLHF, while directly building on tools from RL and language models, is really implicitly impacted and built on theories and philosophies spanning tons of human history. I think we cite Aristotle in this paper, which is fun. It's like going pre-BC, it's like 2,300 years old or something like that. So that's the reason to do this, I think. We kind of list some things in the paper about summarizing what different presumptions of RLHF could be. I think going through these is actually kind of funny. It's fun to talk about these, because they're kind of grab bags of things that you'll see return throughout this podcast that we're talking about it. The core thing of RLHF that, in order to be a believer in this, is that RL actually works. It's like, if you have a reward function, you can optimize it in some way and get a different performance out of it, and you could do this at scale, and you could do this in really complex environments, which is, I don't know how to do that in all the domains. I don't know how to exactly make chat GPT. So it's kind of, we'll overshadow everything. And then there's, go from something kind of obvious like that, and then you read the von Neumann-Morgenstern utility theorem, which is essentially an economic theory that says you can weight different probabilities of different people, which is a theoretical piece of work that is the foundation of utilitarianism, and trying to quantify preferences is crucial to doing any sort of RLHF. And if you look into this, all of these things, there's way more you could go into if you're interested in any of these. So this is kind of like grabbing a few random things, and then kind of similar to that is the Bradley-Terry model, which is the fancy name for the pairwise preferences that everyone is doing. And then all the things that are like, that Anthropic and OpenAI figured out that you can do, which is that you can aggregate preferences from a bunch of different people and different sources. And then when you actually do RLHF, you extract things from that data, and then you train a model that works somehow. And we don't know, there's a lot of complex links there, but if you want to be a believer in doing this at scale, these are the sorts of things that you have to accept as preconditions for doing RLHF. Yeah.Swyx [00:10:09]: You have a nice chart of like the sort of intellectual history of RLHF that we'll send people to refer to either in your paper or in the YouTube video for this podcast. But I like the other slide that you have on like the presumptions that you need to have for RLHF to work. You already mentioned some of those. Which one's underappreciated? Like, this is the first time I've come across the VNM Utility Theorem.Nathan [00:10:29]: Yeah, I know. This is what you get from working with people like to my co-host on the podcast, the rhetoric is that sociologist by training. So he knows all these things and like who the philosophers are that found these different things like utilitarianism. But there's a lot that goes into this. Like essentially there's even economic theories that like there's debate whether or not preferences exist at all. And there's like different types of math you can use with whether or not you actually can model preferences at all. So it's pretty obvious that RLHF is built on the math that thinks that you can actually model any human preference. But this is the sort of thing that's been debated for a long time. So all the work that's here is like, and people hear about in their AI classes. So like Jeremy Bentham, like hedonic calculus and all these things like these are the side of work where people assume that preferences can be measured. And this is like, I don't really know, like, this is what I kind of go on a rant and I say that in RLHF calling things a preference model is a little annoying because there's no inductive bias of what a preference is. It's like if you were to learn a robotic system and you learned a dynamics model, like hopefully that actually mirrors the world in some way of the dynamics. But with a preference model, it's like, Oh my God, I don't know what this model, like I don't know what chat GPT encodes as any sort of preference or what I would want it to be in a fair way. Anthropic has done more work on trying to write these things down. But even like if you look at Claude's constitution, like that doesn't mean the model believes these things. It's just trained to prioritize these things. And that's kind of what the later points I'm looking at, like what RLHF is doing and if it's actually like a repeatable process in the data and in the training, that's just unknown. And we have a long way to go before we understand what this is and the link between preference data and any notion of like writing down a specific value.Alessio [00:12:05]: The disconnect between more sociology work versus computer work already exists, or is it like a recent cross contamination? Because when we had Tri Dao on the podcast, he said FlashAttention came to be because at Hazy they have so much overlap between systems engineer and like deep learning engineers. Is it the same in this field?Nathan [00:12:26]: So I've gone to a couple of workshops for the populations of people who you'd want to include this like R. I think the reason why it's not really talked about is just because the RLHF techniques that people use were built in labs like OpenAI and DeepMind where there are some of these people. These places do a pretty good job of trying to get these people in the door when you compare them to like normal startups. But like they're not bringing in academics from economics, like social choice theory. There's just too much. Like the criticism of this paper that this is based on is like, oh, you're missing these things in RL or at least this decade of RL and it's like it would be literally be bigger than the Sutton and Barto book if you were to include everyone. So it's really hard to include everyone in a principled manner when you're designing this. It's just a good way to understand and improve the communication of what RLHF is and like what is a good reward model for society. It really probably comes down to what an individual wants and it'll probably motivate models to move more in that direction and just be a little bit better about the communication, which is a recurring theme and kind of my work is like I just get frustrated when people say things that don't really make sense, especially when it's going to manipulate individual's values or manipulate the general view of AI or anything like this. So that's kind of why RLHF is so interesting. It's very vague in what it's actually doing while the problem specification is very general.Swyx [00:13:42]: Shall we go to the, I guess, the diagram here on the reinforcement learning basics? Yeah.Nathan [00:13:47]: So reinforcement learning, I kind of mentioned this, it's a trial and error type of system. The diagram and the slides is really this classic thing where you have an agent interacting with an environment. So it's kind of this agent has some input to the environment, which is called the action. The environment returns a state and a reward and that repeats over time and the agent learns based on these states and these rewards that it's seeing and it should learn a policy that makes the rewards go up. That seems pretty simple than if you try to mentally map what this looks like in language, which is that like the language models don't make this easy. I think with the language model, it's very hard to define what an environment is. So if the language model is the policy and it's generating, it's like the environment should be a human, but setting up the infrastructure to take tens of thousands of prompts and generate them and then show them to a human and collect the human responses and then shove that into your training architecture is very far away from working. So we don't really have an environment. We just have a reward model that returns a reward and the state doesn't really exist when you look at it like an RL problem. What happens is the state is a prompt and then you do a completion and then you throw it away and you grab a new prompt. We're really in as an RL researcher, you would think of this as being like you take a state, you get some completion from it and then you look at what that is and you keep kind of iterating on it and all of that isn't here, which is why you'll hear RLHF referred to as bandits problem, which is kind of like you choose one action and then you watch the dynamics play out. There's many more debates that you can have in this. If you get the right RL people in the room, then kind of like this is an RL even when you zoom into what RLHF is doing.Alessio [00:15:22]: Does this change as you think about a chain of thought reasoning and things like that? Like does the state become part of the chain that you're going through?Nathan [00:15:29]: There's work that I've mentioned on one slide called process reward models that essentially rewards each step in the chain of thought reasoning. It doesn't really give the part of interaction, but it does make it a little bit more fine grained where you can think about like calling it at least you have many states from your initial state. That formulation I don't think people have fully settled on. I think there's a bunch of great work out there, like even OpenAI is releasing a lot of this and let's verify step by step is there pretty great paper on the matter. I think in the next year that'll probably get made more concrete by the community on like if you can easily draw out like if chain of thought reasoning is more like RL, we can talk about that more later. That's a kind of a more advanced topic than we probably should spend all the time on.Swyx [00:16:13]: RLHF for decision making. You have a slide here that compares pre-deep RL versus deep RL.Nathan [00:16:19]: This is getting into the history of things, which is showing that the work that people are using now really came from well outside of NLP and it came before deep learning was big. Next up from this paper, Tamer, which is from 2008. Some names that are still really relevant in kind of human centric RL, Bradley Knox and Peter Stone. If you have an agent take an action, you would just have a human give a score from zero to one as a reward rather than having a reward function. And then with that classifier, you can do something with a policy that learns to take actions to maximize that reward. It's a pretty simple setup. It works in simple domains. And then the reason why this is interesting is you compare it to the paper that everyone knows, which is this Paul Christiano et al. Deep Reinforced Learning from Human Preferences paper, which is where they showed that learning from human preferences, you can solve like the basic RL tasks at the time. So various control problems and simulation and this kind of like human preferences approach had higher rewards in some environments than if you just threw RL at the environment that returned a reward. So the preferences thing was you took two trajectories. So in this case, it was like complete trajectories of the agent and the human was labeling which one is better. You can see how this kind of comes to be like the pairwise preferences that are used today that we'll talk about. And there's also a really kind of interesting nugget that is the trajectory that the humans were labeling over has a lot more information than the RL algorithm would see if you just had one state, which is kind of why people think that it's why the performance in this paper was so strong. But I still think that it's surprising that there isn't more RL work of this style happening now. This paper is in 2017. So it's like six years later and I haven't seen things that are exactly similar, but it's a great paper to understand where stuff that's happening now kind of came from.Swyx [00:17:58]: Just on the Christiano paper, you mentioned the performance being strong. I don't remember what results should I have in mind when I think about that paper?Nathan [00:18:04]: It's mostly like if you think about an RL learning curve, which is like on the X axis, you have environment interactions on the Y axis, you have performance. You can think about different like ablation studies of between algorithms. So I think they use like A2C, which I don't even remember what that stands for as their baseline. But if you do the human preference version on a bunch of environments, like the human preference labels, the agent was able to learn faster than if it just learned from the signal from the environment, which means like it's happening because the reward model has more information than the agent would. But like the fact that it can do better, I was like, that's pretty surprising to me because RL algorithms are pretty sensitive. So I was like, okay.Swyx [00:18:41]: It's just one thing I do want to establish as a baseline for our listeners. We are updating all the weights. In some sense, the next token prediction task of training a language model is a form of reinforcement learning. Except that it's not from human feedback. It's just self-supervised learning from a general corpus. There's one distinction which I love, which is that you can actually give negative feedback. Whereas in a general sort of pre-training situation, you cannot. And maybe like the order of magnitude of feedback, like the Likert scale that you're going to talk about, that actually just gives more signal than a typical training process would do in a language model setting. Yeah.Nathan [00:19:15]: I don't think I'm the right person to comment exactly, but like you can make analogies that reinforcement learning is self-supervised learning as well. Like there are a lot of things that will point to that. I don't know whether or not it's a richer signal. I think that could be seen in the results. It's a good thing for people to look into more. As reinforcement learning is so much less compute, like it is a richer signal in terms of its impact. Because if they could do what RLHF is doing at pre-training, they would, but they don't know how to have that effect in like a stable manner. Otherwise everyone would do it.Swyx [00:19:45]: On a practical basis, as someone fine-tuning models, I have often wished for negative fine-tuning, which pretty much doesn't exist in OpenAI land. And it's not the default setup in open-source land.Nathan [00:19:57]: How does this work in like diffusion models and stuff? Because you can give negative prompts to something to like stable diffusion or whatever. It's for guidance.Swyx [00:20:04]: That's for clip guidance.Nathan [00:20:05]: Is that just from like how they prompt it then? I'm just wondering if we could do something similar. It's another tangent.Swyx [00:20:10]: I do want to sort of spell that out for people in case they haven't made the connection between RLHF and the rest of the training process. They might have some familiarity with it.Nathan [00:20:19]: Yeah. The upcoming slides can really dig into this, which is like this in 2018 paper, there was a position paper from a bunch of the same authors from the Christiano paper and from the OpenAI work that everyone knows, which is like, they write a position paper on what a preference reward model could do to solve alignment for agents. That's kind of based on two assumptions. The first assumption is that we can learn user intentions to a sufficiently high accuracy. That doesn't last with me because I don't know what that means. But the second one is pretty telling in the context of RLHF, which is for many tasks we want to solve, evaluation of outcomes is easier than producing the correct behavior. And this is the whole thing. It's like we can compare two poems that the model generates and it can be viewed as liking a positive example, or it could be viewed as really disliking a negative example. And that's what I think a lot of people are doing in like the harm space is like a harmful response to a language model, whether or not you agree with the company's definition of harms is that it's a really bad negative example and they downweight them by preferring something more benign in the RLHF process, among other ways of dealing with safety. So that's a good way of saying it's like this is core, this kind of like comparison and positive or negative example is core to all of the RLHF work that has continued.Swyx [00:21:29]: People often say, I don't know what I want, but I'll know when I see it. This is that expressed in reinforcement learning tools.Nathan [00:21:35]: Yeah, it is. Yeah, it is. That's what everyone's doing in the preference modeling stage that we'll get to. Yeah. Yeah. And you can see there are more papers. This is really just to have all the links for people that go deeper. There's a Ziegler et al. paper in 2019, which shows that you can do this RLHF process on language models. This familiar diagram starts to emerge in 2019, and it's just to show that this goes really far back. I think we can kind of breeze through some of these. And then 2020 is the first open AI experiment that I think caught people's eyes, which is this learning to summarize experiment. It has this three-step process that we'll go to into more when I kind of go into the main concepts. But this is like the first time you see this diagram that they reuse with InstructGPT, they reuse with ChatGPT. And the types of examples that they would have, I don't think I need to read these exactly, but one that I have read a whole bunch of times is like, they took these prompts from Reddit that was like, explain like I'm five or get career advice, and people really pour their heart and soul into these. So these are like multi-paragraph pieces of writing. And then they essentially do comparisons between a vanilla language model, like I think it was either GPT-2 or GPT-3, I don't always get the exact years.Swyx [00:22:42]: 3 was early 2020. So that's about right.Nathan [00:22:45]: Yeah. So this is probably done with GPT-2. It doesn't really matter. But the language model does normal things when you do few shot, which is like it repeats itself. It doesn't have nice text. And what they did is that this was the first time where the language model would generate like pretty nice text from an output. It was restricted to the summarization domain. But I think that I guess this is where I wish I was paying attention more because I would see the paper, but I didn't know to read the language model outputs and kind of understand this qualitative sense of the models very well then. Because you look at the plots in the papers, these Learning to Summarize and Destruct GPT have incredibly pretty plots, just like nicely separated lines with error bars and they're like superfine tuning works, the RL step works. But if you were early to see like how different the language that was written by these models was, I think you could have been early to like things like ChatGPT and knowing RLHF would matter. And now I think the good people know to chat with language models, but not even everyone does this. Like people are still looking at numbers. And I think OpenAI probably figured it out when they were doing this, how important that could be. And then they had years to kind of chisel away at that and that's why they're doing so well now. Yeah.Swyx [00:23:56]: I mean, arguably, you know, it's well known that ChatGPT was kind of an accident that they didn't think it would be that big of a deal. Yeah.Nathan [00:24:02]: So maybe they didn't. Maybe they didn't, but they were getting the proxy that they needed.Swyx [00:24:06]: I've heard off the record from other labs that it was in the air. If OpenAI didn't do it, someone else would have done it. So you've mentioned a couple of other papers that are very seminal to this period. And I love how you say way back when in referring to 2019.Nathan [00:24:19]: It feels like it in my life.Swyx [00:24:21]: So how much should people understand the relationship between RLHF, instruction tuning, PPO, KL divergence, anything like that? Like how would you construct the level of knowledge that people should dive into? What should people know at the high level? And then if people want to dive in deeper, where do they go? Is instruct tuning important here or is that part of the overall process towards modern RLHF?Nathan [00:24:44]: I think for most people, instruction tuning is probably still more important in their day to day life. I think instruction tuning works very well. You can write samples by hand that make sense. You can get the model to learn from them. You could do this with very low compute. It's easy to do almost in like no code solutions at this point. And the loss function is really straightforward. And then if you're interested in RLHF, you can kind of learn from it from a different perspective, which is like how the instruction tuning distribution makes it easier for your RLHF model to learn. There's a lot of details depending on your preference data, if it's close to your instruction model or not, if that matters. But that's really at the RLHF stage. So I think it's nice to segment and just kind of understand what your level of investment and goals are. I think instruction tuning still can do most of what you want to do. And it's like, if you want to think about RLHF, at least before DPO really had taken off at all, it would be like, do you want to have a team of at least like five people if you're really thinking about doing RLHF? I think DPO makes it a little bit easier, but that's still really limited to kind of one data set that everyone's using at this point. Like everyone's using this ultra feedback data set and it boosts AlpacaVal, MTBench, TruthfulQA and like the qualitative model a bit. We don't really know why. It's like, it might just be a data set combined with the method, but you've got to be ready for a bumpy ride if you're wanting to try to do RLHF. I don't really recommend most startups to do it unless it's like going to provide them a clear competitive advantage in their kind of niche, because you're not going to make your model chat GPT like better than OpenAI or anything like that. You've got to accept that there's some exploration there and you might get a vein of benefit in your specific domain, but I'm still like, oh, be careful going into the RLHF can of worms. You probably don't need to.Swyx [00:26:27]: Okay. So there's a bit of a time skip in what you mentioned. DPO is like a couple months old, so we'll leave that towards the end. I think the main result that I think most people talk about at this stage, we're talking about September 2020 and then going into, I guess maybe last year was Vicuña as one of the more interesting applications of instruction tuning that pushed LLAMA1 from, let's say a GPT 3-ish model to a GPT 3.5 model in pure open source with not a lot of resources. I think, I mean, they said something like, you know, they use like under $100 to makeNathan [00:26:58]: this. Yeah. Like instruction tuning can really go a long way. I think the claims of chat GPT level are long overblown in most of the things in open source. I think it's not to say, like Vicuña was a huge step and it's just kind of showing that instruction tuning with the right data will completely change what it feels like to talk with your model. Yeah.Swyx [00:27:19]: From text completion to actually chatting back and forth. Yeah. Yeah.Nathan [00:27:23]: Instruction tuning can be multi-turn. Just having a little bit of data that's like a couple of turns can go a really long way. That was like the story of the whole first part of the year is like people would be surprised by how far you can take instruction tuning on a small model. I think the things that people see now is like the small models don't really handle nuance as well and they could be more repetitive even if they have really good instruction tuning. But if you take that kind of 7 to 70 billion parameter jump, like the instruction tuning at the bigger model is like robustness, little things make more sense. So that's still just with instruction tuning and scale more than anything else.Swyx [00:27:56]: Excellent. Shall we go to technical overview?Nathan [00:27:58]: Yeah. This is kind of where we go through my own version of this like three phase process. You can talk about instruction tuning, which we've talked about a lot. It's funny because all these things, instruction tuning has the fewest slides, even though it's the most practical thing for most people. We could save the debate for like if the big labs still do instruction tuning for later, but that's a coming wave for people. And then like preference data and training and then kind of like what does reinforce learning optimization actually mean? We talk about these sequentially because you really have to be able to do each of them to be able to do the next one. You need to be able to have a model that's chatty or helpful instruction following. Every company has their own word that they like to assign to what instructions mean. And then once you have that, you can collect preference data and do some sort of optimization.Swyx [00:28:39]: When you say word, you mean like angle bracket inst or do you mean something else?Nathan [00:28:42]: Oh, I don't even know what inst means, but just saying like they use their adjective that they like. I think Entropic also like steerable is another one.Swyx [00:28:51]: Just the way they describe it. Yeah.Nathan [00:28:53]: So like instruction tuning, we've covered most of this is really about like you should try to adapt your models to specific needs. It makes models that were only okay, extremely comprehensible. A lot of the times it's where you start to get things like chat templates. So if you want to do system prompts, if you want to ask your model, like act like a pirate, that's one of the ones I always do, which is always funny, but like whatever you like act like a chef, like anything, this is where those types of things that people really know in language models start to get applied. So it's good as a kind of starting point because this chat template is used in our early childhood and all of these things down the line, but it was a basic pointer. It's like, once you see this with instruction tuning, you really know it, which is like you take things like stack overflow where you have a question and an answer. You format that data really nicely. There's much more tricky things that people do, but I still think the vast majority of it is question answer. Please explain this topic to me, generate this thing for me. That hasn't changed that much this year. I think people have just gotten better at scaling up the data that they need. Yeah, this is where this talk will kind of take a whole left turn into more technical detail land. I put a slide with the RLHF objective, which I think is good for people to know. I've started going back to this more, just kind of understand what is trying to happen here and what type of math people could do. I think because of this algorithm, we've mentioned this, it's in the air, direct preference optimization, but everything kind of comes from an equation of trying to learn a policy that maximizes the reward. The reward is some learned metric. A lot can be said about what the reward should be subject to some constraint. The most popular constraint is the KL distraint, which is just a distributional distance. Essentially in language models, that means if you have a completion from your instruction or RLHF model, you can compare that completion to a base model. And looking at the log probs from the model, which are essentially how likely each token is, you can see a rough calculation of the distance between these two models, just as a scalar number. I think what that actually looks like in code, you can look at it. It'd be like a sum of log probs that you get right from the model. It'll look much more simpler than it sounds, but it is just to make the optimization kind of stay on tracks.Make sure it doesn't overfit to the RLHF data. Because we have so little data in RLHF, overfitting is really something that could happen. I think it'll fit to specific features that labelers like to see, that the model likes to generate, punctuation, weird tokens like calculator tokens. It could overfit to anything if it's in the data a lot and it happens to be in a specific format. And the KL constraint prevents that. There's not that much documented work on that, but there's a lot of people that know if you take that away, it just doesn't work at all. I think it's something that people don't focus on too much. But the objective, as I said, it's just kind of, you optimize the reward. The reward is where the human part of this comes in. We'll talk about that next. And then subject to a constraint, don't change the model too much. The real questions are, how do you implement the reward? And then how do you make the reward go up in a meaningful way? So like a preference model, the task is kind of to design a human reward. I think the equation that most of the stuff is based on right now is something called a Bradley-Terry model, which is like a pairwise preference model where you compare two completions and you say which one you like better. I'll show an interface that Anthropic uses here. And the Bradley-Terry model is really a fancy probability between two selections. And what's happening in the math is that you're looking at the probability that the chosen completion, the one you like better, is actually the better completion over the rejected completion. And what these preference models do is they assume this probability is correlated to reward. So if you just sample from this probability, it'll give you a scalar. And then you use that reward later on to signify what piece of text is better. I'm kind of inclined to breeze through the math stuff because otherwise, it's going to be not as good to listen to.Alessio [00:32:49]: I think people want to hear it. I think there's a lot of higher level explanations out there. Yeah.Nathan [00:32:55]: So the real thing is you need to assign a scalar reward of how good a response is. And that's not necessarily that easy to understand. Because if we take back to one of the first works, I mentioned this tamer thing for decision making. People tried that with language models, which is if you have a prompt in a completion and you just have someone rate it from 0 to 10, could you then train a reward model on all of these completions in 0 to 10 ratings and see if you can get chat2BT with that? And the answer is really kind of no. Like a lot of people tried that. It didn't really work. And then that's why they tried this pairwise preference thing. And it happened to work. And this Bradley Terry model comes from the 50s. It's from these fields that I was mentioning earlier. And it's wild how much this happens. I mean, this screenshot I have in the slides is from the DPO paper. I think it might be the appendix. But it's still really around in the literature of what people are doing for RLHF.Alessio [00:33:45]: Yeah.Nathan [00:33:45]: So it's a fun one to know.Swyx [00:33:46]: I'll point out one presumption that this heavily relies on. You mentioned this as part of your six presumptions that we covered earlier, which is that you can aggregate these preferences. This is not exactly true among all humans, right? I have a preference for one thing. You have a preference for a different thing. And actually coming from economics, you mentioned economics earlier. There's a theorem or a name for this called error impossibility, which I'm sure you've come across..Nathan [00:34:07]: It's one of the many kind of things we throw around in the paper.Swyx [00:34:10]: Right. Do we just ignore it?Nathan [00:34:14]: We just, yeah, just aggregate. Yeah. I think the reason this really is done on a deep level is that you're not actually trying to model any contestable preference in this. You're not trying to go into things that are controversial or anything. It's really the notion of preference is trying to stay around correctness and style rather than any meaningful notion of preference. Because otherwise these companies, they don't want to do this at all. I think that's just how it is. And it's like, if you look at what people actually do. So I have a bunch of slides on the feedback interface. And they all publish this.Swyx [00:34:43]: It's always at the appendices of every paper.Nathan [00:34:47]: There's something later on in this talk, which is like, but it's good to mention. And this is when you're doing this preference collection, you write out a very long document of instructions to people that are collecting this data. And it's like, this is the hierarchy of what we want to prioritize. Something amount like factuality, helpfulness, honestness, harmlessness. These are all different things. Every company will rank these in different ways, provide extensive examples. It's like, if you see these two answers, you should select this one and why. And all of this stuff. And then my kind of like head scratching is like, why don't we check if the models actually do these things that we tell the data annotators to collect? But I think it's because it's hard to make that attribution. And it's hard to test if a model is honest and stuff. It would just be nice to understand the kind of causal mechanisms as a researcher or like if our goals are met. But at a simple level, what it boils down to, I have a lot more images than I need. It's like you're having a conversation with an AI, something like type GPT. You get shown two responses or more in some papers, and then you have to choose which one is better. I think something you'll hear a lot in this space is something called a Likert scale. Likert is a name. It's a name for probably some research in economics, decision theory, something. But essentially, it's a type of scale where if you have integers from like one to eight, the middle numbers will represent something close to a tie. And the smallest numbers will represent one model being way better than the other. And the biggest numbers will be like the other models better. So in the case of one to eight, if you're comparing models A to B, if you return a one, if you really liked option A, you return eight if you really like B, and then like a four or five if they were close. There's other ways to collect this data. This one's become really popular. We played with it a bit at Hugging Face. It's hard to use. Filling out this preference data is really hard. You have to read like multiple paragraphs. It's not for me. Some people really like it. I hear I'm like, I can't imagine sitting there and reading AI-generated text and like having to do that for my job. But a lot of these early papers in RLHF have good examples of what was done. The one I have here is from Anthropic's collection demo because it was from slides that I did with Anthropic. But you can look up these in the various papers. It looks like Chat2BT with two responses, and then you have an option to say which one is better. It's nothing crazy. The infrastructure is almost exactly the same, but they just log which one you think is better. I think places like Scale are also really big in this where a lot of the labeler companies will help control like who's doing how many samples. You have multiple people go over the same sample once and like what happens if there's disagreement. I don't really think this disagreement data is used for anything, but it's good to know like what the distribution of prompts is, who's doing it, how many samples you have, controlling the workforce. All of this is very hard. A last thing to add is that a lot of these companies do collect optional metadata. I think the Anthropic example shows a rating of like how good was the prompt or the conversation from good to bad because things matter. Like there's kind of a quadrant of preference data in my mind, which is you're comparing a good answer to a good answer, which is like really interesting signal. And then there's kind of the option of you're comparing a bad answer to a bad answer, which is like you don't want to train your model on two different issues. This is like, we did this at Hugging Base and it was like, our data was like, we don't know if we can use this because a lot of it was just bad answer to bad answer because you're like rushing to try to do this real contract. And then there's also good answer to bad answer, which I think is probably pretty reasonable to include. You just prefer the good one and move on with your life. But those are very different scenarios. I think open AIs of the world are all in good answer, good answer, and have learned to eliminate everything else. But when people try to do this in open source, it's probably like what Open Assistance saw is like, there's just a lot of bad answers in your preference data. And you're like, what do I do with this? Metadata flags can help. I threw in the instruct GPT metadata. You can see how much they collect here. And like everything from the model fails to actually complete the task, hallucinations, different types of offensive or dangerous content, moral judgment, expresses opinion. Like, I don't know exactly if they're doing this now, but you can kind of see why doing RLHF at scale and prioritizing a lot of different endpoints would be hard because these are all things I'd be interested in if I was scaling up a big team to do RLHF and like what is going into the preference data. You do an experiment and you're like, okay, we're going to remove all the data where they said the model hallucinates like just that and then retrain everything. Like, what does that do?Swyx [00:38:59]: Yeah, so hallucination is big, but some of these other metadata categories, and I've seen this in a lot of papers, it's like, does it contain sexual content? Does it express a moral judgment? Does it denigrate a protected class? That kind of stuff, very binary. Should people try to adjust for this at the RLHF layer or should they put it as a pipeline where they have a classifier as a separate model that grades the model output?Nathan [00:39:20]: Do you mean for training or like a deployment? Deployment. I do think that people are doing it at deployment. I think we've seen safety and other things in the RLHF pipeline. Like Lama 2 is famous for kind of having this like helpfulness and safety reward models. Deep in the Gemini report is something that Gemini has like four things, which is like helpfulness, factuality, maybe safety, maybe something else. But places like Anthropic and Chattopadhyay and Bard almost surely have a classifier after, which is like, is this text good? Is this text bad? That's not that surprising, I think, because you could use like a hundred times smaller language model and do much better at filtering than RLHF. But I do think it's still so deeply intertwined with the motivation of RLHF to be for safety that some of these categories still persist. I think that's something I'll kind of settle out, I think.Swyx [00:40:11]: I'm just wondering if it's worth collecting this data for the RLHF purpose, if you're not going to use it in any way, separate model to-Nathan [00:40:18]: Yeah, I don't think OpenAI will collect all of this anymore, but I think for research perspectives, it's very insightful to know, but it's also expensive. So essentially your preference data scales with how many minutes it takes for you to do each task and every button is like, it scales pretty linearly. So it's not cheap stuff.Swyx [00:40:35]: Can we, since you mentioned expensiveness, I think you may have joined one of our spaces back in Lama 2 was released. We had an estimate from you that was something on the order of Lama 2 costs $3 to $6 million to train GPU-wise, and then it was something like $20 to $30 million in preference data. Is that something that's still in the ballpark? I don't need precise numbers.Nathan [00:40:56]: I think it's still a ballpark. I know that the 20 million was off by a factor of four because I was converting from a prompt number to a total data point. So essentially when you do this, if you have multi-turn setting, each turn will be one data point and the Lama 2 paper reports like 1.5 million data points, which could be like 400,000 prompts. So I would say it's still say like 6 to 8 million is safe to say that they're spending, if not more, they're probably also buying other types of data and or throwing out data that they don't like, but it's very comparable to compute costs. But the compute costs listed in the paper always are way lower because all they have to say is like, what does one run cost? But they're running tens or hundreds of runs. So it's like, okay, like... Yeah, it's just kind of a meaningless number. Yeah, the data number would be more interesting.Alessio [00:41:42]: What's the depreciation of this data?Nathan [00:41:46]: It depends on the method. Like some methods, people think that it's more sensitive to the, this is what I was saying. It was like, does the type of instruction tuning you do matter for RLHF? So like, depending on the method, some people are trying to figure out if you need to have like what is called like, this is very confusing. It's called like on policy data, which is like your RLHF data is from your instruction model. I really think people in open source and academics are going to figure out how to use any preference data on any model just because they're scrappy. But there's been an intuition that to do like PPO well and keep improving the model over time and do like what Meta did and what people think that OpenAI does is that you need to collect new preference data to kind of edge the distribution of capabilities forward. So there's a depreciation where like the first batch of data you collect isn't really useful for training the model when you have the fifth batch. We don't really know, but it's a good question. And I do think that if we had all the LLAMA data, we wouldn't know what to do with all of it. Like probably like 20 to 40% would be pretty useful for people, but not the whole data set. Like a lot of it's probably kind of gibberish because they had a lot of data in there.Alessio [00:42:51]: So do you think like the open source community should spend more time figuring out how to reuse the data that we have or like generate more data? I think that's one of the-Nathan [00:43:02]: I think if the people are kind of locked into using synthetic data, people also think that synthetic data is like GPT-4 is more accurate than humans at labeling preferences. So if you look at these diagrams, like humans are about 60 to 70% agreement. And we're like, that's what the models get to. And if humans are about 70% agreement or accuracy, like GPT-4 is like 80%. So it is a bit better, which is like in one way of saying it.Swyx [00:43:24]: Humans don't even agree with humans 50% of the time.Nathan [00:43:27]: Yeah, so like that's the thing. It's like the human disagreement or the lack of accuracy should be like a signal, but how do you incorporate that? It's really tricky to actually do that. I think that people just keep using GPT-4 because it's really cheap. It's one of my like go-to, like I just say this over and over again is like GPT-4 for data generation, all terms and conditions aside because we know OpenAI has this stuff is like very cheap for getting pretty good data compared to compute or salary of any engineer or anything. So it's like tell people to go crazy generating GPT-4 data if you're willing to take the organizational like cloud of should we be doing this? But I think most people have accepted that you kind of do this, especially at individuals. Like they're not gonna come after individuals. I do think more companies should think twice before doing tons of OpenAI outputs. Also just because the data contamination and what it does to your workflow is probably hard to control at scale.Swyx [00:44:21]: And we should just mention at the time of recording, we've seen the first example of OpenAI enforcing their terms of service. ByteDance was caught, reported to be training on GPT-4 data and they got their access to OpenAI revoked. So that was one example.Nathan [00:44:36]: Yeah, I don't expect OpenAI to go too crazy on this cause they're just gonna, there's gonna be so much backlash against them. And like, everyone's gonna do it anyways.Swyx [00:44:46]: And what's at stake here to spell it out is like, okay, that's like cost $10 to collect one data point from a human. It's gonna cost you like a 10th of a cent with OpenAI, right? So like it's just orders of magnitude cheaper. And therefore people-Nathan [00:44:58]: Yeah, and it's like the signal you get from humans is from preferences isn't that high. The signal that you get from humans for instructions is pretty high, but it is also very expensive. So like the human instructions are definitely like by far and away the best ones out there compared to the synthetic data. But I think like the synthetic preferences are just so much easier to get some sort of signal running with and you can work in other, I think people will start working in other goals there between safety and whatever. That's something that's taking off and we'll kind of see that. I think in 2024, at some point, people will start doing things like constitutional AI for preferences, which will be pretty interesting. I think we saw how long it took RLHF to get started in open source. Instruction tuning was like the only thing that was really happening until maybe like August, really. I think Zephyr was the first model that showed success with RLHF in the public, but that's a long time from everyone knowing that it was something that people are interested in to having any like check mark. So I accept that and think the same will happen with constitutional AI. But once people show that you can do it once, they continue to explore.Alessio [00:46:01]: Excellent.Swyx [00:46:01]: Just in the domain of human preference data suppliers, Scale.ai very happily will tell you that they supplied all that data for Lama 2. The other one is probably interesting, LMSYS from Berkeley. What they're running with Chaterina is perhaps a good store of human preference data.Nathan [00:46:17]: Yeah, they released some toxicity data. They, I think, are generally worried about releasing data because they have to process it and make sure everything is safe and they're really lightweight work. I think they're trying to release the preference data. I have, if we make it to evaluation, I'd pretty much say that Chaterina is the best limited evaluation that people have to learn how to use language models. And like, it's very valuable data. They also may share some data with people that they host models from. So like if your model is hosted there and you pay for the hosting, you can get the prompts because you're pointing the endpoint at it and that gets pinged to you and you're any real LLM inference stack saves the prompts tha

As of Jan. 1, 2024, new laws have gone into effect in California regarding traffic stops and enforcement. These laws have implications for pretext stops, racial profiling, and road safety. I explore. *does not constitute legal advice* For more: AB2773, requiring police to explain why they have pulled you over AB645, Speed Safety Systems Pilot Program I also recommended Allow Me to Retort by Elie Mystal if you want to learn more about the seminal case on pretext stops, Whren v. United States. This is the last episode of Season 3. Stay tuned for Season 4, coming soon. :) --- Send in a voice message: https://podcasters.spotify.com/pod/show/natasha-t-baker/message Support this podcast: https://podcasters.spotify.com/pod/show/natasha-t-baker/support

In this episode, Tom gives us a lesson on all things feedback, mostly where our scientific framings of it came from. Together, we link this to RLHF, our previous work in RL, and how we were thinking about agentic ML systems before it was cool.Join us, on another great blast from the past on The Retort!We also have brought you video this week!

A brief introduction to the many problems facing AI and a sneak peak into episode 1, coming soon!

Join our Facebook group by applying here and answering the questions: www.facebook.com/groups/selfperfectedSubscribe here to keep up with our latest podcasts: https://www.youtube.com/channel/UC6nv...Learn more about the group: https://self-perfected.com/Listen to Discussions on reality: https://www.youtube.com/playlist?list...We're on Apple: https://podcasts.apple.com/us/podcast...We're on Spotify: https://open.spotify.com/show/6HXvEho...#SelfPerfectedPodcast #selfperfection

Jonathan Taylor vs. The Colts (0:00) Aaron Rodgers' Retort to Sean Payton (16:21) Brewers Weekend Recap & Deadline Plan (26:31) Matt LaFleur on Training Camp (41:38)See omnystudio.com/listener for privacy information.

In which we read the deepest darkest thoughts of a wrinkly old man seeking a supple young body. Plus, Nomura plays spin doctor in the Ultimania interview! Read the full interview here: https://www.kh13.com/forums/topic/5174-kingdom-hearts-birth-by-sleep-ultimania-interviews/ Drop us a question, comment, or pontification on the nature of the heart! Email us at khbhpodcast@gmail.com --- Send in a voice message: https://podcasters.spotify.com/pod/show/kingdomheartsbyheart/message Support this podcast: https://podcasters.spotify.com/pod/show/kingdomheartsbyheart/support

The Australia-USA connection in the space industry. Apex Space is ramping up. Loft Orbital is staying busy. The SDA is happy with how the Proliferated Warfighter Space Architecture is progressing. SDA Director Tournear is bending ears about supply chain and sustainability. China and Venezuela shake on it, to the moon and back ❤️. And more. Remember to leave us a 5-star rating and review in your favorite podcast app. Audience Survey We want to hear from you! Please complete our wicked fast 4 question survey. It'll help us get better and deliver you the most mission-critical space intel every day. Selected Reading Aerospace Corporation Virgin Orbit Hiring Fair | Event Registration Advanced power systems blaze a trail for America's space effort | GeekWire  SSIB 2023 Hybrid Space Architecture and Advanced Power & Propulsion Working Groups | New Space NM US$400m opportunity for Australian space sector | SpaceConnect Economic opportunities in the space industry | KPMG   Satellite bus startup Apex Space plans first launch aboard SpaceX's Transporter-10 | TechCrunch Loft Orbital orders 15 more buses from Airbus OneWeb Satellites | SpaceNews  Space Development Agency's first satellite launch hailed as model | SpaceNews Tournear: Demand Likely to Grow for Commercial Deorbiting Systems | Via Satellite  SDA's Tournear ‘Just Not' Afraid of Satellite Shootdowns. Supply Chain Is the Greater Worry | Air & Space Forces Magazine China invites Venezuela to join moon base project | SpaceNews  Mars Ingenuity helicopter breaks record for speed and altitude, NASA says | CBS News  Want to hear your company in the show? You too can reach the most influential leaders in the industry. Here's a link to our media kit. Contact us at space@n2k.com to request more info about sponsoring T-Minus. Want to join us for an interview? Please send your interview pitch to space-editor@n2k.com and include your name, affiliation, and topic proposal, and our editor will get back to you for scheduling. T-Minus is a production of N2K Networks, the news to knowledge platform for professionals. © 2023 N2K Networks, Inc. Learn more about your ad choices. Visit megaphone.fm/adchoices

Ed pronounces some Welsh, Brian plays Super Mario, and E contemplates an orb.

Do groundhogs taste better than than those that do not? We talk a wee bit bit about the movie. We also dig deep into the "Do Virgins Taste Better Medley". We can't help but compare what is funny now versus what used to be funny. It's all on the Brobdingnagian Bards Podcast #74   THANK YOU NAGIANS! Brobdingnagian Bards are The Original Celtic Renaissance music duo. They take traditional Irish and Scottish songs and give them a comedic twist. Their unique brand of music on the autoharp, recorder, and mandolin put Austin, Texas on the Celtic music map. Brobdingnagian Bards music is financed entirely by their fans in the Nagians Only Club. For just $5 per month, you can support the Brobdingnagian Bards and help them create new music while enjoying bootleg concerts, videos, and first access to new songs. Join the Nagians Only Club on Patreon today! 0:02 - NAGIANS ONLY: Zingers Recording and editing shows Running your mouth Life's A Faire Faire drama… That 90's show 20:14 - SONG: “Auld Lang Syne” from Real Men Wear Kilts 24:34 - HOW LONG IS A BROBDINGNAGIAN MINUTE? Today's theme: Groundhog Day What's new? What song would you like to hear us talk about? 31:55 - UPCOMING SHOWS APR 1-2, 8-9: Sherwood Forest Faire, Paige, TX APR 6: Dragon Con Filk Music Concert with Brobdingnagian Bards @ 7 PM CST MARC FEB 9: Cat Drinking Songs Concert on Bandcamp @ 7 PM ET MAR 15: St Patrick's Day Internet Concert @ 7 PM ET JUN 3-10: Celtic Invasion Vacations, County Mayo, Ireland ANDREW MONDAYS: Faire Life @ 7 PM CT, Back on Jan 30 39:48 - TODAY'S STORY: 58:03 - SONG: “Do Virgins Taste Better Medley” from A Faire to Remember SONG LYRICS Do Virgins Taste Better words by Randy Farran, music Traditional A dragon has come to our village today. We've asked him to leave, but he won't go away. Now he's talked to our king and they worked out a deal. No homes will he burn and no crops will he steal. Now there is but one catch, we dislike it a bunch. Twice a year he invites him a virgin to lunch. Well, we've no other choice, so the deal we'll respect. But we can't help but wonder and pause to reflect. * Do virgins taste better than those who are not? Are they salty, or sweeter, more juicy or what? Do you savor them slowly? Gulp them down on the spot? Do virgins taste better than those who are not? Now we'd like to be shed you, and many have tried. But no one can get through your thick scaly hide. We hope that some day, some brave knight will come by. 'Cause we can't wait around 'til you're too fat to fly. Now you have such good taste in your women for sure, They always are pretty, they always are pure. But your notion of dining, it makes us all flinch, For your favorite entree is barbecued wench. Now we've found a solution, it works out so neat, If you insist on nothing but virgins to eat. No more will our number ever grow small, We'll simply make sure there's no virgins at all! The Dragon's Retort words by Claire Stephens, music traditional Now, I am a dragon. Please listen to me. For I'm misunderstood to a dreadful degree. This ecology needs my and I know my place. But I'm fighting extinction with all of my race. Well, I came to this village to better my health Which is ever so poor, despite all my wealth. But I get no assistance and no sympathy, Just impertinent questioning shouted at me. * Yes, virgins taste better than those who are not. But my favorite snack mixed with peril is fraught. For my teeth will decay and my trim go to pot. Yes, virgins taste better than those who are not. Well, I'm really quite kind almost all through the year. Vegetarian ways are now mine out of fear. But a birthday needs sweets as I'm sure you'll agree. And barbecued wench tastes like candy to me As it happens our interests are almost the same. You see I'm really quite skillful at managing game. If I ate just your men, would your excess decline? Of course not, the rest would just make better time. Now, the number of babies a woman can bare Has limits. That's why my prunings done there. And an orphan's a sad sight and so when I much. I'm careful to eat only virgins for lunch The Brobdingnagian Bards Podcast was produced by Marc Gunn, and Andrew McKee. Sign up to our mailing list to download free MP3s and get monthly updates of what's new. Find it all at thebards.net

For our end-of year show, we are featuring a couple of our favorite book segments from 2022. First, a Black guy's guide to the Constitution: Elie Mystal explains why “our constitution is not good.” He's The Nation's justice correspondent and author of Allow Me to Retort.Also: “Bad Mexicans”—that's what the revolutionaries of 1910 were called as they fought on both sides of the US-Mexico border against the robber barons and their political allies. UCLA historian Kelly Lytle Hernandez tells that story and talks about her book on race, empire, and revolution in the borderlands.Subscribe to The Nation to support all of our podcasts: thenation.com/podcastsubscribe.Advertising Inquiries: https://redcircle.com/brandsPrivacy & Opt-Out: https://redcircle.com/privacy

On Tuesday's Rugby Daily, Richie McCormack hears from Dan Sheehan and Caelan Doris ahead of Saturday's test with the Wallabies. Dan McKellar's disappointment at Mack Hansen's defection. Gerry Thornley gives us his insight into Connacht's life post-Friend. Plus, Rassie Erasmus says his tweets are NSFR.

The 3rd and final coppice focussed episode is with Duncan and Sam from The Coppice Co-op. We are predominantly talking about Charcoal in this episode, its history, its uses and the ways to make it.This episode is part of a trio focusing on the technique of coppicing.  They were made possible by the Bill Hogarth Memorial Apprenticeship Trust (BHMAT).*** BHMAT are now inviting applications for our next placement and hope to interview and start the placement soon (by the end of the year/beginning of 23 hopefully). *** Episode LinksThe Coppice Co-op - http://coppicecoop.co.uk/wp/ Book - Rebecca Oaks - Making Charcoal and Biochar - https://www.hive.co.uk/Product/Rebecca-Oaks/Making-Charcoal-and-Biochar--A-comprehensive-guide/21833770Local coppice charcoal supplier - https://ncfed.org.uk/public/products/charcoal/suppliers/BHMAT (Bill Hogarth Memorial Apprenticeship Trust) - http://www.coppiceapprentice.org.uk/Hookway Retort - https://hookwayretort.co.uk/Exeter Retort - https://www.exetercharcoal.co.uk/Celebrate The Holiday's Alcohol FreeLearn why this time of year is the best and easiest time to quit drinking .Listen on: Apple Podcasts SpotifySupport the show

Follow Us:https://twitter.com/seenanowhttps://www.instagram.com/justin_williams_comedy/https://twitter.com/arielleatyFollow the show join the discord:Discord: https://discord.gg/WRZ8zgusPTWebsite: https://fraudsters.fm/Twitter: https://twitter.com/fraudstersLPNInstagram: https://www.instagram.com/fraudsterslpn/YouTube: https://www.youtube.com/channel/UCQwl8sDTVEAxhwJdYgm-yrgIntro Music Credit and Sound Fx:https://twitter.com/gograntgordonSeason Cover Art by:https://comedyartwork.com/ Sources and References in Episode: PBS History of Abortionhttps://www.youtube.com/watch?v=PB-zXDYCrzEElie Mystal, “Allow me Retort” https://www.amazon.com/Allow-Me-Retort-Black-Constitution/dp/B09TCGZSDN/ref=sr_1_1?keywords=allow+me+to+retort+by+elie+mystal&qid=1664911698&qu=eyJxc2MiOiIxLjg4IiwicXNhIjoiMS42MiIsInFzcCI6IjEuNzUifQ%3D%3D&s=books&sprefix=allow+me+t%2Cstripbooks%2C120&sr=1-1Report on CPCshttps://prochoice.org/wp-content/uploads/cpc_report.pdfYale Study on Reproductive Justicehttps://openyls.law.yale.edu/bitstream/handle/20.500.13051/7124/Resuscitating_the_Black_Body_Reproductive_Justice.pdf?sequence=2&isAllowed=yPro-Life Journalhttps://web.archive.org/web/20200831230458/https://humanlifereview.com/wp-content/uploads/2015/11/2012-summer.pdfGallup Polling on Abortion https://news.gallup.com/poll/1576/abortion.aspxJoint Statement of Elected Prosecutorshttps://fairandjustprosecution.org/wp-content/uploads/2022/06/FJP-Post-Dobbs-Abortion-Joint-Statement.pdf

Last week, we aired an interview with Jeremy Courtney, who along with his wife Jessica, founded the charitable organization Preemptive Love. We had a lot to say about the controversy surrounding the Courtney's dismissal from the charity that they had spent well over a decade building, but we thought it best to let people hear from Jeremy, first. This week, we got together with Jeremiah to discuss the accusations made against them, the tone and inflection of the articles written by two former employees, the preemptive critiques of Humanite (a new project that has only just launched and is already under assault), and final thoughts on what this mess says about the people involved, both directly and indirectly. This conversation might ruffle some feathers, but it needs to be had. The callous, enthusiastic way in which we discard imperfect people who have done so much good is infuriating. The Courtneys aren't perfect, but they've been in the field helping save lives and build communities for a long time. We think they deserved better than they got. Here are the articles we discuss:   “Preemptive Love: What Happens When a Charity Runs More Like a Cult” by Ben Irwin “Jeremy and Jessica Courtney's misconduct at Preemptive Love, documented” by Ben Irwin “A Donor's Guide to Preemptive Love's Financial Statements” by Ben Irwin “What you should know about Humanite, Jeremy and Jessica Courtney's new organization” by Ben Irwin “An Open Letter to Preemptive Love” by Courtney Christenson The Board of Directors' Statements & Timeline of Events       

Chris is on vacation this week, so we're sharing a favorite recent episode. Please note that this conversation was originally recorded in May of 2022 before the Supreme Court overturned Roe v. Wade. From the original description: “Forced labor is already unconstitutional and what is forced birth other than forcing a woman to labor against her will?” remarked Elie Mystal, a justice correspondent at The Nation, following the leak of a Supreme Court draft opinion that would overturn Roe v. Wade. Mystal is also author of the New York Times bestselling book, “Allow Me to Retort: A Black Guy's Guide to the Constitution,” in which he points out problems with-and solutions for- reversing systemic issues created by America's founding document. He joins WITHpod to discuss his objections to conservative interpretations of rights, abortion rights law, changes he'd make to the Constitution, and revisions he'd make to the structure of the Supreme Court and more.

This week: Allison presents your questions to Elie Mystal about Allow Me To Retort Follow: Elie Mystal https://twitter.com/elienyc https://www.thenation.com/authors/elie-mystal/ Buy the book: https://www.barnesandnoble.com/w/allow-me-to-retort-elie-mystal/1138652220 Follow AG on Twitter: Dr. Allison Gill https://twitter.com/allisongill https://twitter.com/MuellerSheWrote https://twitter.com/dailybeanspod

This week: Allison presents your questions to Elie Mystal about Allow Me To Retort Follow: Elie Mystal https://twitter.com/elienyc https://www.thenation.com/authors/elie-mystal/ Buy the book: https://www.barnesandnoble.com/w/allow-me-to-retort-elie-mystal/1138652220 Follow AG on Twitter: Dr. Allison Gill https://twitter.com/allisongill https://twitter.com/MuellerSheWrote https://twitter.com/dailybeanspod

This week: Allison dives into Chapters 19, 20, and 21 of Allow Me to Retort by Elie Mystal Follow: Elie Mystal https://twitter.com/elienyc https://www.thenation.com/authors/elie-mystal/ Buy the book: https://www.barnesandnoble.com/w/allow-me-to-retort-elie-mystal/1138652220 Follow AG on Twitter: Dr. Allison Gill https://twitter.com/allisongill https://twitter.com/MuellerSheWrote https://twitter.com/dailybeanspod

This week: Allison dives into Chapters 19, 20, and 21 of Allow Me to Retort by Elie Mystal Follow: Elie Mystal https://twitter.com/elienyc https://www.thenation.com/authors/elie-mystal/ Buy the book: https://www.barnesandnoble.com/w/allow-me-to-retort-elie-mystal/1138652220 Follow AG on Twitter: Dr. Allison Gill https://twitter.com/allisongill https://twitter.com/MuellerSheWrote https://twitter.com/dailybeanspod

This week: Allison dives into Chapters 16, 17 & 18 of Allow Me to Retort by Elie Mystal Follow: Elie Mystal https://twitter.com/elienyc https://www.thenation.com/authors/elie-mystal/ Buy the book: https://www.barnesandnoble.com/w/allow-me-to-retort-elie-mystal/1138652220 Follow AG on Twitter: Dr. Allison Gill https://twitter.com/allisongill https://twitter.com/MuellerSheWrote https://twitter.com/dailybeanspod

This week: Allison dives into Chapters 13 and 15 of Allow Me to Retort by Elie Mystal Follow: Elie Mystal https://twitter.com/elienyc https://www.thenation.com/authors/elie-mystal/ Buy the book: https://www.barnesandnoble.com/w/allow-me-to-retort-elie-mystal/1138652220 Follow AG on Twitter: Dr. Allison Gill https://twitter.com/allisongill https://twitter.com/MuellerSheWrote https://twitter.com/dailybeanspod

This week: Allison dives into Chapters 11 and 12 of Allow Me to Retort by Elie Mystal Follow: Elie Mystal https://twitter.com/elienyc https://www.thenation.com/authors/elie-mystal/ Buy the book: https://www.barnesandnoble.com/w/allow-me-to-retort-elie-mystal/1138652220 Follow AG on Twitter: Dr. Allison Gill https://twitter.com/allisongill https://twitter.com/MuellerSheWrote https://twitter.com/dailybeanspod Submit a question for the author: https://formfaca.de/sm/aZjsUltdT Promo codes: Thanks, Ana Luisa. Get 10% off at shop.analuisa.com/bookclub - code bookclub Thanks, Dipsea. Get a 30 day free trial when you go to DipseaStories.com/ag.

Melissa interviews Elie Mystal about his new book, Allow Me To Retort: A Black Guy's Guide to the Constitution.P.S. Melissa, Kate, and Leah will be on The Problem with Jon Stewart this Thursday, June 9th! Don't miss it.

This week: Allison dives into Chapters 7 through 10 of Allow Me to Retort by Elie Mystal Follow: Elie Mystal https://twitter.com/elienyc https://www.thenation.com/authors/elie-mystal/ Buy the book: https://www.barnesandnoble.com/w/allow-me-to-retort-elie-mystal/1138652220 Follow AG on Twitter: Dr. Allison Gill https://twitter.com/allisongill https://twitter.com/MuellerSheWrote https://twitter.com/dailybeanspod Want to support the show and get it ad-free and early? https://dailybeans.supercast.tech/ Or https://patreon.com/thedailybeans Promo codes: Thanks, Ana Luisa. Get 10% off at shop.analuisa.com/bookclub - code bookclub Go to creditkarma.com/loanoffers to find the loan for you.

This week: Allison dives into the intro and Chapters 3 through 6 - of Allow Me to Retort by Elie Mystal Follow: Elie Mystal https://twitter.com/elienyc https://www.thenation.com/authors/elie-mystal/ Buy the book: https://www.barnesandnoble.com/w/allow-me-to-retort-elie-mystal/1138652220 Follow AG on Twitter: Dr. Allison Gill https://twitter.com/allisongill https://twitter.com/MuellerSheWrote https://twitter.com/dailybeanspod Submit a question for the author: https://formfaca.de/sm/aZjsUltdT

This week: Allison dives into the intro and Chapter 1 of Allow Me to Retort by Elie Mystal Follow: Elie Mystal https://twitter.com/elienyc https://www.thenation.com/authors/elie-mystal/ Buy the book: https://www.barnesandnoble.com/w/allow-me-to-retort-elie-mystal/1138652220 Follow AG on Twitter: Dr. Allison Gill https://twitter.com/allisongill https://twitter.com/MuellerSheWrote https://twitter.com/dailybeanspod Promo codes: Get 20% off all furniture and home decor, free shipping on furniture, early product access and much more with a JKH membership. Join at JenniKayne.com/home Thank you Credit Karma. Head to credit karma.com/loanoffers to see personalized offers.

Samantha Bee sits down with author, lawyer, and commentator Elie Mystal to talk about his new book "Allow Me to Retort," going to law school with Tom Cotton, the need to get rid of the Electoral College, what rights SCOTUS might take away next, and Neil Gorsuch's ouija board. Original release date: April 19, 2022.

This week: Allison talks with Elie Mystal as she starts his book Allow Me to Retort Follow: Elie Mystal https://twitter.com/elienyc https://www.thenation.com/authors/elie-mystal/ Buy the book: https://www.barnesandnoble.com/w/allow-me-to-retort-elie-mystal/1138652220 Follow AG on Twitter: Dr. Allison Gill https://twitter.com/allisongill https://twitter.com/MuellerSheWrote https://twitter.com/dailybeanspod Promo codes: Get 20% off all furniture and home decor, free shipping on furniture, early product access and much more with a JKH membership. Join at JenniKayne.com/home Thank you Credit Karma. Head to credit karma.com/loanoffers to see personalized offers.

“Forced labor is already unconstitutional and what is forced birth other than forcing a woman to labor against her will?” remarked Elie Mystal, a justice correspondent at The Nation, following the leak of a Supreme Court draft opinion that would overturn Roe v. Wade. Mystal is also author of the New York Times bestselling book, “Allow Me to Retort: A Black Guy's Guide to the Constitution,” in which he points out problems with-and solutions for- reversing systemic issues created by America's founding document. He joins WITHpod to discuss his objections to conservative interpretations of rights, abortion rights law, changes he'd make to the Constitution, and revisions he'd make to the structure of the Supreme Court and more.

Samantha Bee sits down with author, lawyer, and commentator Elie Mystal to talk about his new book "Allow Me to Retort," going to law school with Tom Cotton, the need to get rid of the Electoral College, what rights SCOTUS might take away next, and Neil Gorsuch's ouija board.

Mary Trump demolishes Ron DeSantis' chances in 2024 and the media hype supporting it.  Then, she welcomes the justice correspondent for The Nation, Elie Mystal, to discuss the critical need for the Democrats to embrace the Black community and take action to rally the base going into the midterm elections.  They also explore whether our country has come far enough on issues of equality and if our Constitution is really all that it's cracked up to be. ‘Ask Mary Anything' Email: MARY@POLITICON.COM Get More From Elie Mystal Twitter | The Nation | Above The Law Blog | Author of “Allow Me To Retort”    Get More From Mary Trump: Twitter | Substack | YouTube | Author of “The Reckoning” & “Too Much and Never Enough”