Captaining IMO Gold, Deep Think, On-Policy RL, Feeling the AGI in Singapore — Yi Tay

25:45Stuff like that. I didn't work on the inference time scale. Yeah. I wouldn't know here. That is mostly that old. And then there was this the codename was apparently IMOCAT, which you named after your desk. Okay, that's not really, like it was in the, so I think I tweeted about it at some point, right? Yeah. Going up to tweet. So the IMO cat was basically like, okay, it's not like an official codename or something. It's just like the name that the config of the job was like IMO cat. That's the. You just need some kind of name. Yeah, I mean, I just like, you know, I just, I like cats and then, yeah.

26:20Yeah, fair enough. That is mostly it on IMO unless you want to bring up anything else. we have other sort of researchy topics, but beyond, before I go into sort of researchy topics, I did want to maybe leave the floor to cover what else should people know about the reasoning effort that's going on at GDM? Let me think of where to start, please. What do people need to know? You know, it's very good, yeah. That's what people need to know. Maybe an easy one to start with would be a lot of people were focusing on maybe academic benchmarks two years ago, last year, maybe LLM Arena, this year Pokemon.

26:53Pokemon is very interesting reasoning, visual reasoning and just general long horizon agent planning benchmark. And I don't know, you seem to focus on it a lot and I think Gemini did very well so obviously I think it's something that is easy to talk about. I think I'll probably should be with this and there's actually nothing specifically done for Pokemon. Of course. Yeah, of course. There's nothing specifically done for Pokemon and I think that I think Logan had this tweet recently about the recent Gemini tree on Pokemon Crystal and Pokemon Crystal is like so much more special yeah I think Pokemon is like so I used to play a lot Pokemon and I'm a big Pokemon fan in general and I think it's a it's a great like you said it's a great long horizon benchmark and stuff like that and I think it's good to check in once in a while on these benchmarks that like almost never get contaminated or like people actually like don't spend time to like you climb benchmarks it's like kind of silly to like people like okay like what are you working on if some people are oh I'm working on Amy, I'm working on HRE, I'm working on Pokemon Maxing or something like that.

27:54That's kind of like funny. We did interview the Cloud-based Pokemon AI. I think his name is David. And it showed serious flaws in Anthropix's screen understanding vision capabilities. I literally couldn't tell. I'm trying to like get past this wall, but you just keep running into it. Because it doesn't know that the wall is there. And so it doesn't have any spatial reasoning at all. I mean, some of it could be like a harness, like the harness thing, or like also whether the model has access to like game state information or is it compared to visual. Yeah, Cloud's implementation is very game state heavy.

28:29They dumped effectively like all the memory of what's going on in the emulator. Yeah, yeah, I see. Yeah. I think for, I don't know whether I'm jumping off tangent or something like that. I think solving Pokemon is going to be more of like how fast you solve it. And then like the thing that I have not really seen so far is like whether the model can complete the Pokédex. Why is that? I don't think it's more challenging. No, completely the Pokédex is so hard. You need to plan, you need to like, you need to search up like information. Like, there's some things if you don't like go online and basically you need to like have a little bit of deep research in this here.

29:02The model will just never know like that it needs to trade. Okay, if it's able to go online, post on forums and then find someone to, like, hey, can I trade with you? Is Pokémon to evolve? Some Pokémon needs to be traded to be involved. Yes, or mated. Yeah, but anyway, I don't know. I have not seen the model available to complete the Pokemon. But complete the Pokemon is really hard actually for models. So I think that's actually an interesting, like, like an interesting one. Yeah. Yeah. I wonder what the real world analogy would be once, if let's say we have a model that is capable of doing that, what can we make it do that we cannot do today?

29:37But there's a lot of planning involved. It's just real deep research. Like planning, there's also a lot of planning involved. And it's more, I think complete the Pokemon game is very linear, right? The thing with Pogodex involves a lot of like... Backtracking. Research. Yeah, a lot of research and a lot of... Yeah, so it's probably a different nature itself. Is that as interesting to you as, for example, a lot of other people in the AI for Science world are trying to discover things that you cannot look up, right? Noble knowledge. Yeah, noble knowledge. Because basically what you're saying is we're not even there yet.

30:08We're at the place where models cannot consistently apply knowledge that they look up, right? You give Gemini access to web search and you say, okay, come try to collect all the Pokemon in a Pokedex. You don't have high confidence that it will do it. I don't know if someone's actually tried. Probably not, right? No, I think the hard part is actually trying to synthesize the web knowledge and then apply it in the game itself with all that visual state going on and stuff like that. It probably would be solved in one or two. It is challenging, but it's not super interesting. You're basically just...

30:42The task really is, can you look up the guide to do it? And then can you apply the guide? That's it. You know what's even more intelligent than that? Creating the guide. Like being the first to figure out how to create the guide. Which is what it is. Oh yeah, but then when it comes to this, it's mostly just an exhaustive search thing. A model will just try and try, like humans and guides for that. Okay, so that's actually less interesting to you. Interesting. Okay, actually, when you think about it, it's not super, super interesting, but it's okay. It's just like I have not seen a model to try to do this.

31:11Yeah, I think efficient search, of a novel idea space is interesting. Obviously, you can brute force anything, but we're not talking about brute forcing. We're talking about trying to create an AI scientist. But novel knowledge is actually an interesting thing that I think is going to be quite a big thing. Being able to generate novel... Google has done stuff there which I don't... You're probably not that close to those teams that has done AI scientist work. There's some things that have been... It's like, for example, if you freeze the model weights at like 2015, you freeze time at 2015, and then you, even with the current model, let's say you have, okay, let's assume there's no leaking of information somehow.

31:49Can, if you ask the model, what's the best ML? Okay, not 2015, like 2012 or something. It would just tell you like SVMs are like the best, right? This is the way that machine learning works in general, right? And then the question is, can you invent the transformer? Right? It might not be able to, like even today models, they might not even be able to invent the transformer. Like if you freeze the time at a certain time and even you bring the tech, I mean, the model is a transformer, so I just say there's no, assuming there's no leakage. issue. No, it's probably possible. So I think there's still a lot of open questions about whether the model can really innovate and generate really novel knowledge.

32:23Yeah. Yeah. One related question on that, which I think is related to the Danny paper, which is I think people have this sort of mythicism on what reasoning is. And reasoning, if you really demystify it a lot, it's whatever happens inside the chain of thought tags, right? and you post, you're eliciting that reasoning behavior from some stuff that is already latent inside of the pre-trained corpus. That's one version of this interpretation. I think these days, like reasoning itself is very vague and it's very open. So it's like mostly different people will have different definitions of what reasoning is, right?

33:00So I agree that this chain of thought is like basically when people think of reasoning, they associate it with chain of thought and obviously it's what happens in the thinking and then, okay, reasoning, right? But I think these days it's more like, I said in earlier, but it's like reasoning in RL is almost like this, like basically it's anything that is like post-training to at least capabilities, basically. It's like RL and post-training to at least capabilities. So I think the actual like technical definition of reasoning is making models better with thinking and post-training. Okay, yeah. So basically like RLing the model to think better, right?

33:34And thinking is more like thinking tracers and thought trajectories and stuff. like that right there's also this line of work for late like latent thinking and stuff like that like when the latent thinking and discrete token thinking is like going to be the same thing or like something like that it's like an open question meaning adding extra tokens to your vocab that represent all there's all this i don't forgot what's the name like these academy papers that do these loopy things or track token or like they basically instead of decoding discrete tokens you actually simulate this by doing this in latent space right so when you do chain of thought thinking and like reasoning you basically decode extra tokens hide in the thinking tag and then you decode stuff but like latent thing is basically you just don't decode tokens you just don't bother by it it might start speaking the native language of thinking is numbers not passing it through some filter of english and sometimes you must start thinking chinese or something else yeah generally i'm not i'm not really i don't really believe that model thoughts have to be the same with human thoughts i'm actually like generally in ml i'm more of the school of thought of let the model do whatever it wants in general there was a discussion there's a latent representation hypothesis paper that i think you're maybe sympathetic to if you haven't already read it to me sounds obvious basically image models will have the same idea what a laptop is versus a text model will have they converge on like the same latent yeah and obviously you can align them and you can do all those stuff with them and so it totally makes sense that their concept would be just a vector of numbers that represents laptop that's the concept yeah and And okay, maybe you have some numerical differences between one model's idea of what a laptop is versus another, but it mostly would be the same.

35:10Yeah, very interesting. The question I was kind of leading into was that because there's now we're in this age where LLM text is in the corpus of stuff that we train on, where it's a little bit of a recursive loop, right? Like the reasoning tokens are out there now. And so pre-trained models themselves, preaching base models that are also capable of reasoning. And they're increasingly so as more and more reasoning text goes into the corpus. Isn't that interesting? Or is that worrying? Do you actually see much reasoning traits on the internet? I've never seen those though. I would say that... Like on Hug Your Face?

35:49Yeah, people are publishing that specifically. As to whether or not people, researchers are actually including that in their training corpuses, who knows, right? Like, that's their choice. But I would say that percentage on Common Crawl that has COT tokens in there went from 0 to 0.001%. And it will just go up over time because people are publishing it. Yeah, but I think if the sources are quite clear, you can actually filter away those because usually people put it on GitHub. Do you want to filter it? Maybe you don't. There's a choice for the... Yeah, quite literally. the whole reason why I don't think we covered this in our previous part, but two years ago, a lot of people were like, oh, you just include more coding tokens in your pre-trained corpus.

36:34But the coding tokens are different from like coding tokens. It generalizes outside of code for reasoning. Oh, that was like... You don't believe that? No, no, no. I don't know if it's still true today, but I see. Yeah, that was just our general coverage of reasoning. I would say that there's a lot of interesting work here and more to do. Maybe I'll cover one thing which I know that you have personal inputs on, which is that you have started using AI coding. Oh yeah, so I actually don't really use much AI coding in the past, but I think we've reached a point where AI coding has started to become really useful.

37:08Like, okay, so before AI coding, the thing that I find the most useful about these models in general is when I have these big spreadsheets of a lot of results and I just need plots of it. I think models can quite go to the screenshot and make a plot of this. I hate making this method. talk like stuff about is so annoying okay but that's basically like one thing that i can remember about like how i use ai in the past but i think ai coding has started to become the point where i run a job i get a bug i almost don't look at the bug i place it into like anti-gravity and like i throw it that would fix the bug for me and then i relaunched the job that like beyond like vibe coding is more like vibe training vibe ml or something like that i would say it does pretty well most of the time and it's actually there are classes of problems that is just generally i know this is actually really good for and in fact maybe probably better than like i would have to spend 20 minutes to find like figure out the issue and then fix the thing and then we don't so yeah that's very interesting because i would say like level one vibe coding is you actually know what to do you're just too lazy yeah it's just i just do it for me like i've done this a thousand times like just go fix it like i know exactly what to do here you're saying it's like like the next level where you actually don't even know it's investigating it for you as long as like the answer looks right you'll just strip it at the start I was a bit like I did check it look at the thing and then at some point I'm like okay maybe the model looks better than me so I'm just going to let it do its stuff and then I will relaunch the job based on the fix that the model gave me and I think the models will just keep getting better and better so yeah it's something that I also think that recently there's this anti-gravity I think also because these tools were not like that in google infrastructure is not that easy to you don't i'm not that familiar what it's available outside and when i was in the startup i didn't really i think the models were not like so good like one and a half years ago so it's also like a forcing function that like it's also people like oh try adding gravity is a game changer and stuff like okay so i just started using and yeah yeah you spent some time with varoon recently what did what did well no i really did say hi and i guess you were telling me that you're an ai researcher that doesn't even use much AI and like now you're actually like AI-pilled as a user.

39:22There were so many moments this year where AI suddenly crossed that like that immersion thing. I think AI coding is one of them that we just discussed. I think like Nano Banana also got to the point where I usually like if you make these images it's just like for fun it just troll your friend or something like that. But like Nano Banana actually really got so good that you can use it for charts. That you can use it for like basically yeah so it's getting really good and I think yeah this year the stuff like And even things like the past, many of these LMS will like hallucinate things a lot. But now I just trust it automatically.

39:53I think we just, people are just enjoying the utility by these models. So now I'm like, I was always AI fuel. AI is a good thing. I don't see how anybody can disagree with that. Yeah, but you are actually using it for things that you are high expertise in, which is your own ML work. Yeah, yeah, yeah. And just to come back, do you have a special version of Gemini that you use internally that we don't have access to or it's like public Gemini? I think it's the public Gemini. Okay. Yeah. I was just saying like, it would be entirely reasonable to train Gemini internal for only your code base and your work.

40:31Oh, actually, I'm not sure though. You see, these things are just like, it's rather way for me. But obviously, if it obviously improves your productivity by, I don't know, 10%, yeah, worth it, right? Yeah. So I think that's interesting. And there's the interesting thing, levels of how much do you trust it? How much of your jobs do you automate away and you no longer need? There's also the question I guess about how people come up and train in the field if they you no longer need juniors. Because Gemini is your junior ML researcher. So I think these are all interesting questions. I want to say one quick thing first, right?

41:04So I think that when it comes to like whether a model can be like a junior suite or like something like that, right? So I think if you think of it at this way of if a job from a one one X suite, one time suite, can be replaced by a model itself. But let's say you are a manager, right? The objective that, the metric you track is like your time. And then if you can have a model that saves you like the same amount of time as the work that your reports do, but you don't actually like replace one person per se, but you, yeah. A little bit from everybody. Yeah, right. Then you can, I definitely agree that, okay, like when you count the net time save, there are times where the model can fix bugs that like would have cost me like one day, right?

41:41One day is huge. And these things are definitely like, I don't know whether anybody has done any real metric evaluation or this type of things, but if you use time as a real metric and then not as a number of, okay, maybe three hours kind of metric, but these things are not going to replace one person as it is, but more like a passive aura that buffs everybody. The in-game terms, right? I often think of myself as a bard because I tell stories and I plus everybody around me. That's an ideal situation for me in a D &D group. Oh, okay. I don't play D &D, but okay. I get it. They're the kings of Passamora.

42:17Support heroes. Support, support. Yes. Yeah, okay. AI support, I think, is very encouraging. I think like, where is it still not working for you? That you've tried and you're like, oh man, I expected it to be better. There are times when models try to get lazy and try to fix something. They are still... We won't have... They get lazy and then they try to like, gaslight me into thinking that like, the bug is fixed. so there are still classes of there are classes of problems that are like very easy for the model very hard for humans there are some things that are very easy for humans very hard for modern and stuff so it's still very hard to characterize these these things into this proper quadrants and stuff like that so i would say that the capabilities of models these days are good enough to be like really helpful but like it's still you is a bit like it still has some but yeah but i think this world like this i don't think there's anything that to be done to specifically like focus fire These things is more like general capability improvements.

43:12The model just gets better over time and then these things will just like go away. You say that, okay, so yes, I think obviously in the grand scheme of things, just trust the process, keep scaling in every dimension and things will just fall away, things will emerge. But you've also said in the past, I can't remember the exact tweet, where you were like, each additional data set compounds over time. They're just small additions. And I would say that when you say things like focus fire on things that you would think humans, it's easy for humans, hard for machines. Those are easy wins where you can just add a dataset that would focus fire on that.

43:46And isn't hill climbing just a sequence of doing that until you reach AGI? Okay, so I get your point. I think that it's true that sometimes a lot of progress on the whole is just a series of small incremental changes that push. I think that's accurate. That's true. There's also, it also feels that there's also a lot of like small, like seemingly minor for the lack of better word, like that push AI to the state where it is today. So I definitely agree. So nothing against people who like focus fire, but it's just that when I mean that like, it might be not easy to focus fire on things that are like not very easy to characterize.

44:22So it's just that like when he has, there's something targeted, right? Like, okay, I want to improve this capability, add some data. So I think like defining like, if I was defining the problems and stuff like that, it's like characterizing it. And if it can be characterized, then okay, then fine. But I think it's like, what I was trying to say with the coding is that these things are not even, some of the class of problems are like, I don't work on coding, but like people maybe work on coding, they know like, maybe they have like terminology for different types of failures. But so maybe somewhere somebody is focusing fire on this, then make the model better.

44:52That's great for everybody. Yeah, I mean, that's why it takes a thousand people to like get all these things together. AI is definitely like a big collective effort these days. It's a big machine. Yes, it's really crazy. Okay, so I just wanted to broaden out to general things people are talking about in the community on research. Which again, I know that you are very locked in, so you don't necessarily have read all the papers or anything, but we can just riff on it. You can obviously ask me what I think as well. Is attention all you need? So attention and transform is like a core idea in the recent times.

45:27like pre-training and scale is the thing that made attention and transformers like actually shine right because without i think the first transformer paper was this like a machine translation thing and then basically gpt and bird were the ones that like actually showed the full like big potential of this idea so in terms of is attention like really really all we need like probably no but i think it's like one of the from the architectural point of view also maybe maybe know, but it's not all you need. But you need it definitely. So what else are you thinking about on that same level? Are you talking about MOE stuff?

46:03What do you mean when you say it's not all you need? You definitely need the scale of pre-training. You need all the tokens. You need like RL. I think when I say, when people say, I say it's the potential you need, it's mostly from an architectural point of view. Will transferbers get us all the way to AGI, right? I guess it's the... So basically when you get to AGI, it's the problem. Will it still be a GNOME architecture or like meaningfully different? It would be a transformer, I think. Really? It depends on what you call it, but I think unless the paradigm shifts completely, which as a scientist, you cannot completely say no to that.

46:37This would never happen. But my feeling is that it's been almost 10 years since the transformer. 2017. 2009 years since the transformer. I think we have not replaced self-retention. It's some form of it. You could rename it. you could name it something else you could sometimes you can do local global yeah it's still a transformer in the end and I think like that's not going like anywhere unless the whole thing with like back prop like everything like goes like the whole thing just changes completely like then there's a different story there's a different conversation to have but if it's still within the same scope and bounds so I spent a lot of time thinking of about architectures and like whether there's alternate architectures and stuff okay at the sequence processing level like there is the ultimate sequence to sequence transformer.

47:26It's probably the self-attention is there was this whole big era which I was also involved in this era where people try to like undermine the attention as much as possible like they try to remove it simplify it make it efficient like this whole like efficient attention era and at the end of the day the outcome was always like oh we remove all the attention but we have one layer of self-attention there and it still works like that's at the end like always a story. Which even Noam a character he published some stuff about how he has some ratio of mixing of local and global attention, right? Like, basically, still attention, but modifying it quite a lot.

48:01I will continue local and global attention to be like, spill attention. Just like how much you're skipping. Yeah, the only question is that if the formulation changes too much, your QKV becomes like ABCD, EFG, or something like that. Okay, maybe I'll give you some motivating constraints in order to do this. You guys are still charging 2X for over 180k token context or 240, something like that. And the max, theoretical max is 2 million tokens, right? What if we need 200 million? Is there some point at which where even this concept of input token context is irrelevant because you are doing continual learning, that kind of stuff, where you're modeling it as, okay, the AGI will be achieved through a sequence-to-sequence transformation.

48:47Therefore, an attention is the best sequence-to-sequence model or architecture, therefore attention is all you need. But I think other people are like, sequence to sequence doesn't accurately capture intelligence. But that's not really about sequence to sequence. It's more about like the whole gradient design and backprop thing, right? It's not architecture itself. That's a problem that we, that is more of like the learning paradigm itself rather than the architecture itself. I think the architecture is just basically like the interface between the learning algorithm and the tokens. I think it's more about the learning algorithm itself.

49:20And this, like, continuing learning this, like, there's many ways to think about processing many, like, insanely large, like, contexts, right? Like, 200 million, 1 billion tokens or something like that, right? Like, whether it's going to be, like, you have a new learning algorithm that every time you run inference, you learn on it, right? Then you can technically have some kind of memory, like human beings learning as I'm talking to you, right? So that's also, like, one way. The other way is, like, whether, okay, maybe somebody will say that, okay, the attention is, like, it's too expensive for 200 million, one billion contacts, so we need new architecture.

49:53Or some people will say that, oh, we just improved the chips, like accelerators. So I think many ways to interpret it, but I think it's like, if it's about, there's a lot of fundamental things that, if it's about continual learning and stuff, there's a lot of fundamental things about the learning algorithm and stuff as well. That will have to change? I think like the learning paradigm and architecture and stuff, that goes like hand in hand. And I think as the field progress and ideas just stack on top one another, right? So there's also this thing about the idea that it was proposed has to be compatible with all the work that has been done before to shine, right?

50:24It's a bit of a variant of this hardware lottery like by Sarah. It's not the hardware lottery that I wrote about the GPUs failing, but it's the original hardware lottery, but it's more like a bit of a lottery of like the things proposed have to play well with the things that were proposed before. So it's a bit like going down this local opt-in minima to some extent. So now we are like in this local minima of like transformers and everything, everything, right? Maybe it's not easy to like get totally out of this. Because also a lot of people's investment optimization have been done. So the things that play well need to play well with the ideas before.

50:58And the way I see it now is it's very difficult to like come out of it. Okay, I'm not entirely convinced. I see what you're saying, but let's call it Gen AI. I fucking hate that term. It's still a very young few. And so, yes, there's been eight years of work on the Transformer. But what's that in the grand scheme of things? maybe we're in a local minima and we've got to notch ourselves out of it I do want to leave that open-ended I don't have an idea, I do think that people are in what Ilya Suskiver has been calling the age of research, right? We're like, okay, we scaled up what we can scale up we know what the next maybe 1, 2 orders of magnitude look like in scaling on every dimension that we know about, but what is the next dimension to scale?

51:41There's this misunderstanding misunderstanding a little bit about like all the last five years has just been like scaling things. Scale. Okay, please tell me more. Yes. You made that joke about now we scale researcher salaries. Okay, let's not go there. But I think that ideas like matter and I think that there have been a lot of good ideas in the last five years. It's just that maybe it's just not, so it's not been like blindly, like if you took an MLP, right, just like without self-addiction and you just, okay, I'm going to throw like$100 trillion on it and scale up that thing and the thing is never going to work.

52:15It's never going to work. Yeah, yeah. So there's no, like, there's part of it that's also, like, I think the bitter lesson gets used too much in, like, too conveniently used around. But actually, there's also a little bit of, not a bitter, there's also a sweet lesson where it's like, ideas matter. And I think even to today, right, like, people downplay ideas and stuff like that. Yeah. Do you think the rate of new ideas, without being specific about what ideas, because obviously you can't share, but like, do you think the rate of ideas has increased or decreased because there's like, kind of a law of definition returns?

52:44Are they smaller? I think the number of ideas is always proportional to the number of researchers working on a certain problem. So by definition, it should increase. But I think the number of ideas that actually work is not decreasing compared to the last. We're not in the era of diminishing returns yet. So I think the ideas are still very important and there's still very good ideas that are game changers that are being invented. And I think I know the answer to this, but is the closed lab advantage increasing versus open source or decreasing like the chinese labs they say keep publishing open source models and some of the american labs as well publish open source models would you say that the ideas that i see there nvidia has nemotron opening i has gpt oss these are all basically checkpoints on what's publicly known about training models as of this year you You know, it's declassified information because everyone does this.

53:44I think that the gap is increasing. I don't think it's completely predictable from stuff that you said before. I think the gap is definitely increasing. Yeah. I think that justifies researchers. Otherwise, what's the point of having researchers if not finding new tricks that compound over time? Yeah. Yeah. But definitely, I think it's increasing. Yeah. Okay. I'll do a side tangent. I don't know if you have any comments on this. So then this is very related to NVIDIA's recent purchase of Grok, which I don't know if you have views because you're very TPU-centric, but are we memory or compute bound?

54:19And this is relevant to the Transformers discussion of like... In terms of what, like serving? Exactly. I think the classic view is that we're compute bound because we just need more compute for pre-trained and RL and then inference. but actually the counter argument I would make against this is I actually have these charts of Moore's laws I wish I could just pull it up easily Moore's laws of the scaling of compute versus scaling of memory versus scaling of network and bandwidth and compute has a much higher slope of scaling than the other two memory I mean what the chip memory like honestly I don't think about this this memory bound that much so maybe it doesn't

55:28So I would disagree with it But honestly, I don't think about a dozen years. Well, it's this, that, that, that, that much. Yeah, I understand. Okay, data efficiency. So this is a joke, but implicit in this is that there's some kind of a maximum data exposure, right? And so I, so previously, I would say that a lot of the training paradigms is like one epoch is all you need. It's the mini title of this idea. I would say that the real number maybe is between three to four epochs. and I do wonder what the theoretical limit of data efficiency of a model in terms of training and compression should be.

56:05I don't know what that means. Data efficiency and basically but you're asking the question in a way of asking how much repeats is tolerable? Is that... One, tolerable is contingent on does it actually improve in meaning? It's not about you actually want to do it for its own sake. But I do think there's that and then there's also just the sheer amount of stuff that we can learn with limited data. So you say, let's say you're not compute bound, you're not memory bound. But let's say you are data bound. Last time that we were on the podcast, we talked about chinchilla versus inference optimal training.

56:39But now actually, I think a lot of people are even talking about data optimal training. Like given limited data set, how well can you learn from it? I think that's an interesting research direction that not enough people are talking about. Maybe it is something that is commonplace in the labs, but it seems very clear that we are very unoptimized with regards to how much we learn from our data. I'll just put it there. I think in general, like learning more, like extracting more from varied data points is definitely valuable, but I think that also relates to the fact that we're running out of tokens in the world.

57:19so I don't work on data for pre-training and I think things that I say were general state of industry not nothing the general state of industry so I think that I don't even know whether data has diverged the way that these things are done it has diverged too much across these labs and no open there's a lot of cross-pollination for sure cross-pollination? okay, okay but I don't think about data that much like the pre-training data that much yeah Maybe earlier this, the first half of this year, I would have said that kind of pre-training is dead. And that everyone's like just funneling all their work towards RL.

57:56And we had this like Grok chart, which is very interesting, where we're sending the same amount of compute on Cycran as to, you think it's a Cy-O. No, I don't know. I have no idea. I think that people are taking it seriously. They are like, yeah, okay, whatever. Especially in the agent labs, like Cognition Cursor, they're taking the open source models from from whoever and then adding let's call it pre-trained scale rl on top of it if they have that level of info which data which they do which is very interesting i would say yeah this data efficiency argument yeah i think to me is also more trying to discover new paradigms of learning in order to get where we want to all go which is yeah and the existence proof is humans, right?

58:41Your two-year-old daughter is much more capable than an LLM in some things having seen way, like, eight orders of magnitude, less data. That's very interesting. Yeah, comparing human learning and machine learning is definitely like... Purely as an existence proof that we could probably do better. Three examples of dog. Yeah. Fourth example of unidentified animal, I can probably tell is a dog, as a human. But machines, classically, you take 20. The data efficiency for off-humans is definitely way higher than models. Yeah, the only question is that where does this thing come from? Is it actually like putting more flops on every token?

59:19Or like maybe it's like back to the question about whether the transformer is the optimal architecture. Maybe it's backprop. It's the weapon. Maybe it's the off-policy-ness. Maybe it's the... Yeah. So what is the... Like, where is the bug, right? Exactly, exactly. But maybe it's a feature, not a bug, I don't know. So, okay, we've identified and probably it took me a while to get this across. So this is the kind of data efficiency I'm talking about. I think it's emerging. Basically, at the end of every year, I try to take bets as to, okay, what will be the big themes for next year? I think there's more of them that people are really trying to focus on.

59:49Because you're feeling this data crunch, even though everyone's still investing in data, I forgot to mention that I would say that I've been wrong on pre-training being dead. Yes, I've now met pre-trained leads from both Anthropik and Ofnii. And I've seen the talk from the DeepMind guy recently. and so everyone's investing in pre-trained still which is like nice to see. Nobody said pre-trained was dead. I know. No, it's a theory that we're trying to disprove or prove. Anyway, so I think, okay, let me wind back to my general idea, right? So yeah, data efficiency seems worthwhile. You would treat it as like, okay, well, show me where the bug is and I'll go fix it.

1:00:28We don't know where the bug is. We just have existence proof that it could be better. And then I think the final logical chain in this for me is that everyone is focusing on some idea world model as a version of this for more efficient learning, which potentially might not take the form of a sequence-to-sequence transformer. I don't know how that works. Like definitely a little bit out of my depth here. To me, that is more efficient because every world must be internally consistent. And if the next piece of evidence come in and invalidates those worlds, then you no longer need to pursue those paths ever and you can just narrow in on the world that you've identified and so to me that is learning where you're learning to fit world models or to the actual data yes so yes maybe you can treat the learning process as curve fitting yeah so you're learning the world instead of learning the world model yeah i said i'm learning the world model right okay by sampling multiple world models and then finding out which one fits the data the best so i guess my query is this what people talk about is this if you i mean obviously feel free to attack it because i'm just spitballing but this is what i pick up from talking to multiple people about okay what are you talking about with world models what are you talking about data efficiency and learning efficiency and like how do you all together in a cohesive sense of the future where we can actually what's the definition of world model at the start from start yeah there are three kinds okay okay go on yeah first kind is the vo kind or the what's the other one genie that deep mind has which is the sort of video world model.

1:02:01You model everything with some kind of Gaussian splats or whatever, and you inhabit that 3D space. Second, let's call it the Jan LeCun slash meta school of thought, which I don't know if you're that familiar with it. He has published the Jepak architecture, and then separately, Fair has also published the Code World Models, where you're basically, specifically for Code, very interesting, you are executing code and modeling the internal state of the execution environment as you go line by line. The LM actually learns to predict those things. And actually, it seems a lot more efficient at the scale that they've tested it out, which is very cool.

1:02:37Which definition are you anchoring on? The third one. The code world model? That's the second one. Those two are bundled together. The JEPPA fans are probably hating me right now because I'm lumping all metals work under one school of thought. Yeah. But whatever. Okay. The first one is VO, Genie, F.A.L.E.S. Super Spatial Intelligence. Those kind of video-based world models. Second one is some execution or some sort of explicit modeling as you run through the corpus. And I think that the third one is this amorphous thing, which I think people are trying to get to, where they are doing what I said about the resolution of possible worlds and your curve fitting as you learn, as you inference.

1:03:21Yeah, but what is the world model itself? itself is it like it is a mental model of where everything is and how you think the world works how what i think you think like everything okay but technically it's like it is something in the latent space okay okay so you can for simplicity it could be just be like a transformer model between and yes yes so to me that is the most coherent thing to the current paradigm which is you could actually do this in current transformers i think the way that you train it will probably have to be different. Okay. I see, I see. I don't have any conclusion here. I'm just throwing it out as something where I know you're interested in this kind of stuff and I don't have that many knowledgeable people to talk to about it.

1:04:02No, I don't think about world models that often. I think because world models are just not really well defined in the first place. But I think... So don't say world models, but the problem is learning efficiency and maybe, I guess, accuracy or like AGI capability that is not easily unlocked right now on our current path. of scaling yeah so i think i think when it comes to like like data efficiency i think it's more like uh i'm believer of finding ways to spend more flops per token right because you actually basically you're if you are data bound you want higher data efficiency because you can learn more from every data point squeeze squeeze out more points right so things like that can extract more can use more flops on every token is definitely like a form of data efficiency then there's the learning algorithm, right?

1:04:48Because I think there's this, there's a different scaling law for like humans is this, machines is this, like dogs are this, cats are this, there's this different, nice exponent. There's the ALEA chart. Right, yeah, yeah. There's this famous, famous, famous chart. And point one and point two are just like not entirely like the different things because it could be that better architecture is actually just spending more flops per token. So if you are, you come to a point where you are very data bound but not compute bound at all, you just find algorithms that spend a lot of compute on every token on every token.

1:05:18So I think the overarching point is just that, okay, it's a learning algorithm thing for data efficiency and then if whether the correct way is actually just to apply more flops per token just to squeeze out more from every data, every data point. Also because humans actually don't like, like they are exposed less, when you say less or more data it's also very ambiguous because they are technically like on 24, 7 and then you have a lot of like different types of inputs, right? And whether they actually spend more flops on everything that they listen is also a question because maybe they are just better efficient just because, I mean, somebody needs to count like how much flops the brain use to process like how much, maybe they're just spending more compute on every token and also maybe the learning algorithm is different.

1:05:59So, but I agree that data efficiency is very important given that I think we're going to like, there's a limited amount of data in the world. One more thing before we go into DSI. You know how like, we're talking about RL and like, you're working on RL stuff. Why are people paying so much for RL environments? So who is paying for RL environments? OpenAI and Topic at least. Nobody said anything about DeepMind. So a lot of the model labs that are not you are well known for paying at least seven figures for external startups to create RL environments for them to train in. Okay. And I think the question is, if your models are so good at coding, what do you do yourself?

1:06:40And so I think there's some amount of expertise that's being distilled from human experts into an RR environment that you can then let your agents run wild in. But I'm curious if there's any other deeper insight than that because I'm not satisfied with my own explanation. RR environments that have a lot of domain expertise are probably very valuable. And actually, I don't know specifically about what RR environments people are actually buying. But what was the thing that you're not satisfied by? It was so valuable. And a lot of people are saying like, look, it's a next year's app inside of a Docker container that logs stuff out when you send inputs in.

1:07:20Then you can probably do it yourself internally, right? Why you pay so much for some startup that you don't know to do it for you? Actually, I have no clue about why this is happening. Yeah, I have no clue. And a classic example would be like, if you want to build a computer use agent for buying things in e-commerce, you would want our environments that perfectly replicate maybe the top thousand e-commerce websites. and then you just parallel roll out on all of them. Does that seem meaningful? I don't know. Alright, cool. DSI and LLM Rexis. A big bet for me this year for my conference was we actually started focusing on LLM Rexis.

1:07:59The other... Actually, what's the motivation behind starting LLM Rexis? Track. Yeah. I think Rexis is the king AI problem in consumer. It is the single most valuable thing. All your feeds, even search is Rexis. Basically, it's search. basically like it's a retrieval it's the god problem right because rex is ranking but then also filtering also personalization also re-indexing and and like performance it is the god problem and you get paid a lot for it engineers are not that excited by it which is very weird because they don't see a lot of them don't work on rex's and they probably never will yeah but they don't see the monetary value that can come out of a good rex's the other two pieces of updates for me, which I actually didn't even know that DSI directly tied into this, was one Twitter publicly adopted their feed algorithm as an LM Rexus.

1:08:53LM's are just used everywhere now, but whether it's actually like a... Like a big LM. Whether it's like a generative retrieval type of model, it's like another question. Correct. Is it? We don't know. All we know is that they have said that they have swapped out their current Rexus for an LM-based Rexus. that's how they found it. But what is published is YouTube where they actually adopted semantic IDs for YouTube's rexies. And YouTube is obviously a big deal. Is it like public information? Yes. Okay, okay. They came and did talk about it with us. And then they published a V2 this year as well.

1:09:24Okay, I see. So basically, the last time you were on a podcast, we didn't talk about DSI that much. But you have actually some background in IR. You care about IR. I don't care about IR, but I think DSI, okay, like DSI or Genitive Retriever was like I think one of my favorite works in the old of like I have some IR background in like when I was doing a PhD I did some Rexy's work I did some retrieval work with Rexy's and stuff so I have some IR and Rexy's background so I think generally retrieval and generally Rexy's is very very conflicted DSI started as a retrieval thing so we did like natural questions like ranking of documents documents everything it started off as I mean that's actually we did an interview with Yannick, me and Don, we did an interview when the paper came out like a long time ago.

1:10:14So at that time, we wanted to like reimagine retrieval and such. So we wanted, at that time, LAMP, we were still using T5 models at that time. It was like not, we're not in the LAM era yet. It was pre-LAM era. It was like, okay, pre-training works kind of thing. And then there were like some pre-trained models around. So we wanted to reimagine retrieval, right? But retrieval rexists, they're all the same formulation, ranking, retrieval problem, right? And then that's where we started to imagine retrieval as one giant that encodes everything in the memory. We tried so many different, like, semantic ideas, actually, my collaborator, Vin, was the one that came up with either semantic ideas that basically, and the start of this whole genetic retrieval was actually basically literally just trying to give a document like identifier and just predicting like raw brute force predicting like this.

1:10:57It actually works because the models can memorize something. If you look at the literature from all the way to things like Doc to Vag, the words have no meaning in this ID in a vocab it's an obvious number and technically the models have enough capacity to predict but I think semantic ID was an idea that basically you have some semantic association and then you actually try to break down the search space hierarchically so how this work evolved in the RACS was at a time after DSI came out so Atchis Group and Mahesh the guy who they did some exploration of applying DSI to RACSYS. And that's how that generative regenerative RACSYS recommender system paper came out.

1:11:42Yeah, I didn't even know he was involved. It's crazy. That was like, basically us transferring this, like basically, okay, DSI will try to try it on RACSYS. And then I think the recommender system people have a slightly different way of doing semantic IDs, but it's basically just because the domain is slightly different. But after that, I think we were done with the invention part and with this one. The rest is details. The rest are details. So over time, I also left Google and stuff. So over time, these things evolved a little bit here. I think I also saw something like Spotify is also using something like YouTube Spotify.

1:12:16They use this type of semantic IDs, this type of DSI-like models. I think from the research community point of view, the DSI work was the first one that decodes semantic tokens. But then when we went to, I don't know, academic community is strange in a way that they will do things like, oh this is gender retrieval it's not gender rexist it's like they do this kind of random things that is a bit strange but yeah this was the whole history of this gender retrieval apparently there's also a lot of people working on I don't follow actually I don't follow this at all now it's not even in my mind but there was once I went to even in the Singapore office they are Swiss actually like working on gender retrieval they don't work on it I don't know whether they're still working on it but I met a person that tried to explain gender retrieval to me it was quite funny that I kind of co-invented generative retrieval.

1:13:03But I think this whole IRR thing has been just an interesting phase. And I definitely think DSR is one of my more creative works that I've done. That it's not really LM, but it's like... Under the general principle of apply ML to everything, if the Googler is working on generative retrieval, would that be like AI overviews? Is that something similar? I have no idea. Okay. For people listening, I did have a track there. I think you just type in AI.engineer and you'll get it. where the Gemini guy was talking, sorry, the YouTube guy was talking about how to use Gemini or the Rexis. I don't know what size of Gemini because he didn't talk about it.

1:13:40But this is public work now. And basically every YouTube video uploaded gets encoded into some kind of codebook and they retrain this every day on some kind of batch job. Yeah, just interesting. So yeah, I don't know if you even know what Gemini is being used for. I don't follow like this. this day and this day. I do think like in the sense of like for people who are not still not getting it, applying intelligence and the general intelligence of an LLM to the retrieval, to the recommendation task means you can accommodate such weird recommendations, like such weird queries as well that normally like no classical system can ever handle.

1:14:22And I think like it's also somewhat emergent in a sense that when you were using T5, you just couldn't actually add that much value on top of a normal BM25 retrieval technique. Would you say that's accurate? It is not just about paraphrasing, it is about understanding query intent. BM25 is a really strong baseline, actually. BM25 is a really strong baseline, yeah.

1:14:46I don't know the comparative delta versus T5 for you guys versus BM25, but I don't expect it to be very high and I expect it to be a lot higher for a true LLM-based basis, depending obviously on the query set. I didn't really think about it this way before, but because I've done modeling in many different domains, including like search, and even in the search community, I have a community, there's also a benchmark and stuff like that for like that people who climb on and there's some like, there's Amy or like a, I don't know what it's called these days anymore, but generally the modeling dynamics of IRL tasks is very different from, like NREX's tasks is very different from standard language tasks or like vision tasks and stuff.

1:15:27Like when we kill climate LOM, when you train models, like the way that modeling things interact with this environment is very different. So I think that I honestly, I hated working on like Rexxys and retrieval stuff. Okay, I'm just looking about old days. When you work on like T5, you work on, you change architecture, you try to improve complexity, you use superglue, like this is the olden days. You are on like, even now when you train LOMs, you just do zero shots, two shots stuff like that. Your things, because as a researcher, engineer you just interact with the environment a lot by this you're just like okay rl by this environment but rexys and ir has a very strange feeling to it strange feeling in the sense that it feels like you're like whatever works it's like you are in this in a world with the gravity is different or like you are in a world where the modeling things that feel intuitive are not intuitive so it feels like a very strange space to so i wrote some papers back in my days on like Rx's and stuff like that.

1:16:23Every time I ran some modeling experiments for Rx's and stuff like that, I didn't enjoy it. It feels like the environment was rude. It feels like the vibes are just like... What makes it rude? It just feels... Transactional. No, not transactional. I don't know how to describe it. For example, if you play sports like the tennis, where you hit the ball, you have a very nice feeling like hitting the sweet spot. When you do modeling in traditional, when you get the feedback back, you feel like everything sounds right, everything feels right, everything right. But Raxxys and IRR is like a problem where it's like you hit the shutter clock and you hear a glass shuttle like randomly.

1:16:58It just feels this weird sense of war that... Cause and effect are too far apart. Like it just feels strange. And then sometimes maybe the metric like... I think Raxxys, they use like all the NCCG effects. And then the BM25 is strong and then you just... And then you get like worse than like the BM25. Back in the day when you stack 2 LSTM to 3 LSTM, you were like, whoa, I see like... It's just the game. It's an unrewarding area to work in. it's just weird also the IR community and the retriever community is also like always behind the mainstream and then now it's just probably going even more worse because of ILM and stuff so okay I'm getting into Hortick territory but it's just like certain conferences are just like behind NewRibs and ICML and stuff yeah some conferences are just like they're just like applying things that they're downstream they're downstream okay so it always feels very uninspiring to work on on this look there's a reason that you left.

1:17:52But it was like a side quest. Like, I work on it as a side quest thing, yeah. Yeah, okay. Understand. I still think it's an important business problem, even though maybe it's an unrewarding field. You can understand why. Because the academic benchmarks for those tasks are just so far detached from what industry is. I didn't work on any of this like the thing, but that's from an academic point of view. Oh, then all you need are online invals, right? Yeah, yeah. Detests and go. to, ooh, okay. That would have been a different experience, yeah. That is mostly our sort of topic, research topics, coverage and everything.

1:18:27I think we're just going to end on a very simple one on GDM Singapore. You organized a symposium. Here we brought Jeff Dean, Kwok and all the others. Basically, what's the general message or the impetus for starting GDM Singapore? So we'll talk about the event first. So the event was mostly, so Kwok and I are going to start a team and then I think before I came back, We discussed this for some time. Jeff was very supportive of this. He was in the region many times, in Vietnam and Singapore, around the time where I was going to come back. And I think that, so this event itself was, Kwok and Jeff were visiting, and we just inspired the community here.

1:19:07I think that it's also a bit more like a soft, like setting the tone right for the start of the Gemini team in Singapore. And I think it's a very rare instance where you get somebody like Jeff and Kwok, who are the true pioneers of AI in the world to be in one room. And then, are you there as well? And I think like many people told me that - Not true pioneer of AI, but okay. I was there to live tweet. Yeah, I think having them all in one room and then giving these talks, like many people came out to me and said they were very inspired by their presence in the region. So starting a team and starting something is also very, like there's also no one moment that it's not okay.

1:19:43I press the button and it starts, right? It's like a process, right? So we hire people and then people join one by one and stuff, something like that, right? So I think this event was more, I would say, like to set the vibe. I think it's possible for Singapore to be close to the frontier. And I think that we're having the true pioneers of AI here. We want to give this, basically more like an inspiring thing. And also get led for Quark and Jeff to meet the people here. So Jeff was here last year, but Quark hasn't been there for some time and he's going to have a team here. So it's like also nice to bring him around and meet the people here.

1:20:15So it was already like amazing event. We met Long as well and who a lot of people don't know has a CS degree. It's like one of the few PMs with a CS degree. Yeah, yeah. I would say that the context of the meeting was more like partially like also he wanted to learn more about the IMO stuff. Oh, really? And then also about Jeff. Oh, because they invited you without knowing that these guys were coming or something like that. I was a bit like Jeff and Kwok and me, we went to visit the Cherif at the East Tanger and we discussed a little bit on DeepThing, discussed a bit of IMO, and then I think the rest of it was more like Jeff and Lee Seng-ho was talking more about, like, it became less about AI and tech, more about very macro, economical, political thing, which was, I was very out of element with, so I was just, I was just, you were in a suit.

1:21:09I was just talking about the DeepThing and the IMO and stuff like that, but he seemed generally quite surprised that AI has reached this point. I think. But it was a interesting... I would say for people, you have done something that is unique in Singapore's history so far. You're establishing a Frontier Research Lab in Singapore, which is an accomplishment. I think the other thing also that I guess I'm still trying to wrap my head around is, does geography actually matter? You're all working on a team, you have your London people, you have your Mountain View people, and mostly you're just collaborating with them anyway.

1:21:44You've collaborated with them your whole life. I don't even really know what countries mean anymore when it comes to research or just AI in general, because this thing is just inherently international from the start. This is a very good question. Also, it's related to the thing about identity, right? Because I think you also moved from SF in Singapore quite a bit, right? I was in Maldonville just like one, two weeks ago. And I'm here, but almost all my... If you just look at my siblings, aside from my family, like everybody I talk to is like somehow in the Bay Area or like that's because of work and everything I think the geography matters okay firstly the most like thing is like logistical is like probably the time zone so you literally want the 24 hour coverage around the world no there are there are advantage no what I'm saying is that the difference is like it's also more like like people define the location more than the location define okay the people somehow okay the time zone we get to the time zone a little bit like the pros and cons I think that's pros and cons, right?

1:22:44So you're bullish on Asia, Singapore people, the talent pool? I think we managed to find like amazing people. But I also have to say that this type of things is more like talent attract talent. I think most of the time, like people are very excited. Like the vibe I get is that people are very excited because it's like Coase team and my team. And we're working on library core things related to AGI. So I feel like the talent we can get from the region is really good. But it's only because it's us. We can unlock this talent. Otherwise, we might join some other place. Yes, and move to the US. Yeah, yeah, yeah.

1:23:25About the identity-wise, I would say that I definitely agree with you that why it doesn't matter. I think the advantages of Singapore itself or just anywhere, it's also that you are like... Okay, the world is very good. So technically, you can interact as much as you want. You can also go there. But I think Singapore has this advantage where you can go close and you can go far. I think the Bay Area is like so much about... I have friends in London and New York that would just never move to the Bay Area. I'm not against Bay Area. I think Bay Area is a great place, right? But it's just AI, AI, AI, AI, AI, everywhere.

1:23:59Right, I think sometimes if you have some like mental space and energy to have some other culture and then like, you know, like London, Singapore, New York, they have their own culture, right? But the Bay Area culture is just like AI, right? Like you just go anywhere, you just hear AI everywhere, right? Even the billboards and stuff. You can hear a bit much. Yeah. Although I did see some billboards down here. Or AI also? Yeah, I was like, what is this? It's a culture infecting Singapore. I do think that to some extent, if you want to do research in, you need a little bit of peace and quiet somewhere, right?

1:24:32So this island may be good for that, but then you can, you're still like able to like be connected right okay so i think that's mainly talent wise i think people are strong here uh yeah so far enough away but you're still connected you have strong talent what are you hiring for you're still hiring right we're hiring like like my team will work on like rl and reasoning for gemini and gemini deep thing i think we care more about like talent density now so we're not also growing that big, there's more diverse just because compute per capita is probably important. So I think that's something that we're hiring for now.

1:25:13Basically, I personally, there's a lot of the ends, but I think generally there's a lot of people who are very capable. But I think what I'm looking for mainly is either you have a track record of RL research or some, even not necessarily the RL, or like some exceptional achievement in like coding competitions or like some exceptional achievement somewhere, then there's like the kind of people that we want. Yeah, because you don't strictly require... I do remember something about your record days where you're like, you like to train your own trainers from scratch, right? So you don't... I forgot if I said that or not, but to some extent, to some extent, I think we'll definitely be very happy with people that are like very high stats and just like even without much statistical knowledge?

1:26:00No, no, stats. Yeah, yeah. The stat points. Child with int. Like just high tech tech points people, like just raw IQ, high tech talent people like this. I think all like strong engineering skills, ML, ML can be learned easily. Our knowledge can be learned easily. Yeah. I think maybe one version of this is, can it be done on a student budget? Can you do something interesting anymore on a student budget? I would say relevant to the point where conferences are quieter these days. I did do an interview with one of the best paper winners where they worked on thousand layer neural network RL. And that was done on the student budget.

1:26:36It was very cleanly executed pieces and paper and good findings. Look, I'm not sure if production models will ever go to a thousand layers, but they stretched it in an interesting direction and found some good recommendations. And the guy immediately got hired by opening eye. and I think that's encouraging for the grad students in the market who are like, okay, well, do I need to know somebody who works at these labs in order to get in? My uncle works there. I get the internship or whatever. No, actually, you can just do it on a student budget with good advisors. Oh, I actually think one thing interesting is that for most of the people that I actually went to recruit them like personally, right?

1:27:10So you see the work and then you send them DMs, right? So I get a lot of best people. No, no, no. Like for hiring generally. So generally, I almost, to your point, it's like you almost like you can just do good work, put it online and then somebody will contact you, right? It's actually super easy but super hard at the same time because I can tell you like I talked to a few of these grad students, they don't know what good work means, right? Because they don't know there's so many things their professors have the agenda they're forcing on them which like may not be right because it's not like their professors know what to work on either.

1:27:44So yeah, they just need guidance. They just need to, hey, work on these like five things. It showed me an interesting result in any of them. okay so if somebody comes up with something and then does something that you feel that is very tasteful and it aligns with what like researchers in the labs like like one and they come up with that independently you know that the function that produce is good right like if you just go and tell somebody to do this like you you can you just get the signal that this guy can execute right so i think there's some value in people that yeah they demonstrate taste research taste yeah research taste.

1:28:15Very interesting. I feel like I could give people, yeah, I do care about this. In some ways, the research directions work that I do is a little bit of that. Like, it's low accountability for me because obviously it's just thought experiments. But I think for a lot of people, it's like their career is bounded by can you demonstrate research taste with this like short three, four years that you have and just do it. Yeah. Yeah, I would say that this is more, there's so much competition just because of like the, like everybody wants to get into AI. It's just more of like how to, like mostly it's more like how you're going to prove yourself.

1:28:50Yeah. It must be hard these days to be a grad student trying to prove yourself. It's definitely harder, but yeah. Yeah, not your job. Okay, that was it. Do you have any other sort of rants or topics that you had queued up before we wrap? I don't, yeah. But it was great. It was, I had a great time. It was fun chatting, man. Fun, fun chatting. Yeah, even, I even love, last time we were supposed to do, last time we meet at the symposium, we were supposed to record. But even if we just ended up hanging out and chatting, it's just nice to get the brain dump of what's going on in your world. Because yeah, we're working on really important stuff, man.

1:29:19Always great to chat with you and see you here. Good to chat. Parting words on the sort of weight loss and workout journey. Because there's also a big thing for you. I think being healthy is important to do good research. Right. And I think I've been probably in one of... I'm probably in the peak of physical health now. Yeah, you look great. Yeah, thanks. and I think it's also impacted my work in a good way. You did the sort of Kabat-y inspired like biohacking. I didn't go so extreme but I was like also quite data-driven when I came I would like have my own email and then I would track this. I'm still supposed to make a blog post about this but I feel like I'm not like really at the end game yet so like when I get there I will but yeah just to just for people who don't know I think I lost 23 kilos this year actually across one year yeah one and a half years yeah so 23 yeah basically literally from the last podcast to now yeah there's an abolition study now 23 kilos yeah and I think like my HRV heart rate variability has went up by two times and my rising heart rate has dropped by 30 beats per minute 30 beats per minute like it was like 80-90 and now it's 160 oh yeah 80-90 is super high yeah I was unhealthy yeah yeah yeah okay yeah so I think when it's hard like what do you have the thing that kept you going, you know.

1:30:41A lot of people, they maybe, they're focused on AI, including myself, right? I do prioritize work. I enjoy work. I don't enjoy the fitness side. But obviously, it feeds in to your intellectual work, like the sort of log off and go for a walk, eat better, all that kind of stuff. It obviously feeds in. But like people seeing a positive example like you, they will get inspired to do the same thing. So I think it's good to set yourself up as an example. I think definitely helps. When I do these things for my health, I just think that it's also part of work because it helps me to get better at my job.

1:31:14So it's important as well. I think it's important as well. I like the HIV off the bat. I have no idea what mine is, but yeah. There's a general question about what is productivity and how do you measure it? What really matters? And it's still unclear to me. But I do think general energy level and hunger almost. like you almost have to like experience physical hunger in order to have intellectual hunger and I don't know if that's like a thing no when I'm hungry I just think of food I think to me it's like it's destroying things but it's hard to do work when you're hungry yeah okay thank you so much yeah thanks it's really great yeah have a great time

1:32:04Thank you.