Speaker unverified: evaluation

Transcript context

…in the training data for all of the language models. And so the language [00:32:00] models already know these repos. In some cases, the language models already know the individual issues in Swebench. So basically, like, some of the training data has leaked. And so it, it definitely will overestimate with respect to that. I don't think it's like, you know, Horribly, horribly off but I think, you know, it's boosting the accuracy by a little bit. So, maybe that's the biggest reason why. In terms of asking for help, and whether we're benchmarking asking for help yes we are. So one one thing we're working on now, which we're hoping to put out soon, is we we basically made SuperVig. Sweep edge issues. Like I'm having a, I'm having a problem with the matrix multiply. Please help. Because these are like, if anybody's run a popular open source, like framework, these are what half your issues are. You're like users show up and say like, my screen doesn't work. What, what's wrong or something. And so then you need to ask them questions and how to reproduce. So yeah, we're, we're, we're working on [00:33:00] that. I think. It, my impression is that agents are not very good at asking for help, even Claude. So like when, when they ask for help, they'll ask for help when they don't need it. And then won't ask for help when they do need it. So this is definitely like an issue, I think. Yeah, so great question. There's a lot of work on web [00:34:00] agents. I didn't go into, like, all of the details, but I think there's There's three main ways that agents interact with websites. The first way is the simplest way and the newest way, but it doesn't work very well, which is you take a screenshot of the website and then you click on a particular pixel value on the website. And Like models are not very good at that at the moment. Like they'll misclick. There was this thing about how like clawed computer use started like looking at pictures of Yellowstone national park or something like this. I don't know if you heard about this anecdote, but like people were like, oh, it's so human, it's looking for vacation. And it was like, no, it probably just misclicked on the wrong pixels and accidentally clicked on an ad. So like this is the simplest way. The second simplest way. You take the HTML and you basically identify elements in the HTML. You don't use any vision whatsoever. And then you say, okay, I want to click on this element. I want to enter text [00:35:00] in this element or something like that. But HTML is too huge. So it actually, it usually gets condensed down into something called an accessibility tree, which was made for screen readers for visually impaired people. And So that's another way. And then the third way is kind of a hybrid where you present the screenshot, but you also present like a textual summary of the output. readers for visually impaired people. And So that's another way. And then the third way is kind of a hybrid where you present the screenshot, but you also present like a textual summary of the output. And that's the one that I think will probably work best. What we're using is we're just using text at the moment. And that's just an implementation issue that we haven't implemented the. Visual stuff yet, but that's kind of like we're working on it now. Another thing that I should point out is we actually have two modalities for web browsing. Very recently we implemented this. And the reason why is because if you want to interact with full websites you will need to click on all of the elements or have the ability to click on all of the elements. But most of our work that we need websites for is just web browsing and like gathering information. So we have another modality where we convert all of it to markdown because that's like way more concise and easier for the agent to deal with. And then [00:36:00] can we create an index specifically for agents, maybe a markdown index or something like that would be, you know, would make sense. Oh, how would I make a successor to Swebench? So I mean, the first thing is there's like live code bench, which live code bench is basically continuously updating to make sure it doesn't leak into language model training data. That's easy to do for Swebench because it comes from real websites and those real websites are getting new issues all the time. So you could just do it on the same benchmarks that they have there. There's also like a pretty large number of things covering various coding tasks. So like, for example, Swebunch is mainly fixing issues, but there's also like documentation, there's generating tests that actually test the functionality that you want. And there there was a paper by a student at CMU on generating tests and stuff like that. So I feel like. Swebench is one piece of the puzzle, but you could also have like 10 different other tasks and then you could have like a composite [00:37:00] benchmark where you test all of these abilities, not just that particular one. Well, lots, lots of other things too, but Yeah, so that's a great question. So we have a basic coding and browsing agent. And I won't say basic, like it's a good, you know, it's a good agent, but it does coding and browsing. And it has instructions about how to do coding and browsing. That is enough for most things. Especially given a strong language model that has a lot of background knowledge about how to solve different types of tasks and how to use different APIs and stuff like that. We do have [00:38:00] a mechanism for something called micro agents. And micro agents are basically something that gets added to the prompt when a trigger is triggered. Right now it's very, very rudimentary. It's like if you detect the word GitHub anywhere, you get instructions about how to interact with GitHub, like use the API and don't browse. rigger is triggered. Right now it's very, very rudimentary. It's like if you detect the word GitHub anywhere, you get instructions about how to interact with GitHub, like use the API and don't browse. Also another one that I just added is for NPM, the like JavaScript package manager. And NPM, when it runs and it hits a failure, it Like hits in interactive terminals where it says, would you like to quit? Yep. Enter yes. And if that does it, it like stalls our agent for the time out until like two minutes. So like I added a new microagent whenever it started using NPM, it would Like get instructions about how to not use interactive terminal and stuff like that. So that's our current solution. Honestly, I like it a lot. It's simple. It's easy to maintain. It works really well and stuff like that. But I think there is a world where you would want something more complex than that. Yeah, I think the Anth, so the Anthropic MCP is like, a way to It, it's essentially a collection of APIs that you can use to interact with different things on the internet. I, I think it's not a bad idea, but it, it's like, there's a few things that bug me a little bit about it. It's like we already have an API for GitHub, so why do we need an MCP for GitHub? Right. You know, like GitHub has an API, the GitHub API is evolving. We can look up the GitHub API documentation. So it seems like kind of duplicated a little bit. And also they have a setting where [00:40:00] it's like you have to spin up a server to serve your GitHub stuff. And you have to spin up a server to serve your like, you know, other stuff. And so I think it makes, it makes sense if you really care about like separation of concerns and security and like other things like this, but right now we haven't seen, we haven't seen that. To have a lot more value than interacting directly with the tools that are already provided. And that kind of goes into my general philosophy, which is we're already developing things for programmers. You know, how is an agent different than from a programmer? And it is different, obviously, you know, like agents are different from programmers, but they're not that different at this point. So we can kind of interact with the interfaces we create for, for programmers. Yeah. I might change my mind later though. So we'll see. It's definitely not random. It's like, if you think it's more complex than it's. Like, just intuitively, it's more likely to fail. I've gotten a bit better at prompting also, so like, just to give an example it, it will sometimes fail to fix a GitHub workflow because it will not look at the GitHub workflow and understand what the GitHub workflow is doing before it solves the problem. So I, I think actually probably the biggest thing that it fails at is, um, er, that our, our agent plus Claude fails at is insufficient information gathering before trying to solve the task. And so if you provide all, if you provide instructions that it should do information [00:42:00] gathering beforehand, it tends to do well.…

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