Shawn Wang: prediction

Transcript context

…Yeah. Yeah. I like drawing references to other things that are happening in, in the, in the wild. So, 'cause we only get to release this podcast once a week. Mm-Hmm. , the LAMA three paper also had some really interesting. Thoughts on synthetic data for code? I don't know if you have reviewed that. I'll highlight the back translation section. Because one of your dataset focuses is updating documentation. I think that translation between natural language, English versus code, and back and forth, I think is actually a really ripe source of synthetic data. And Llama3 specifically called out that they trained on that. We should have gone more into that in our podcast with them, but we, uh, we didn't, we didn't know, but, uh, there's a lot of interesting work on synthetic data stuff. We do have to wrap up soon, but I'm going to briefly touch on the submission process for SuiteBench. So, you have a 30 percent state of the art SuiteBench result, but it's not on the leaderboard because of submission issues. I don't know if you want to comment on, on, like, that stuff versus, uh, you know, we also have, like, we also want to talk about SuiteBench verified. Um, yeah, just anything on the benchmarking side. The potted history of this is, is, is quite simple, actually. SweeBench, up until, I want to say two weeks ago, but it might be less than that, or more than that. But I think two weeks ago, suddenly started mandating what they call trajectories, when you submit. So, but prior to this, essentially, when you run SweeBench, you run it through their harness, and out the other end you get a report. json, which is like, here's how many I resolved, here's how many I didn't resolve, these are the IDs, the ones I did, these ones the IDs I didn't, and it gives you any ones that might, might have errored, or something like that. And what you would submit would be all of your model patches that you outputted and that report. And then you would like PR that into the sweep entry per and that would be it. That was the still the case when we made our submission on whatever day it was. They look at them every Monday. We submitted it at some point during the week. I want to say it was for four days before that. And, um, I sort of like sat back and waited. I assumed it would be fine when it came to Monday. Um, they then said, actually, no, we want model trajectories. And I was like, okay, let me see what this is. And so on. I sort of dug into it and like model the trajectories are essentially the context window or like the reasoning process of like, show you're working. How did you get here? If you do a math exam, show me you're working. Whereas before they were like, just give me the final answer. Now they want to see the working, which I completely understand why they want to see that. Like the SWE bench fundamentally is an academic research project and they want all the stuff to be open source and public so people can learn from each other and improve and so on and on. Very good. I completely agree. However, at least for us, and the reason that we are not on the leaderboard is that obviously the model outputs that we generate are sort of a mirror of our training data set, right? Like you train the model to do a certain thing and output a certain way. Whatever your output looks like, your training data for the moment, as a closed source company, like fighting for an Edge, we've decided not to publish that information for that exact reason. I don't want someone basically taking my tra. And then taking a model that's soon going to be GA and just distilling it immediately and then having genie for themselves. And, you know, as a business owner, that's the decision I've had to make. The patches are still public. So like the, dare I say, traditional SweeBench submission, you can go to our GitHub repo and see it and run them for yourself and verify that the numbers come out correctly. Like that is all, that is the potted reason. That's the story. That's the story. Uh, SweeBench verified. You have a score. I do have a score. I do have a score. 43. 8%?…

Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.