31–40 of 209 posts
Jokes aside, a benchmark I look forward to is ARC-AGI-3. I tried out their human simulation, and it feels very reasoning heavy.
Leaderboard: https://arcprize.org/leaderboard
(Most premier models don't even pass 5 percent.)
> We audited a 27.6% subset of the dataset that models often failed to solve and found that at least 59.4% of the audited problems have flawed test cases that reject functionally correct submissions, despite our best efforts in improving on this in the initial creation of SWE-bench Verified. Is this saying a quarter* of the questions and answers were wrong, this whole time?! If so, how was this ever, in any way, a va…
Imagenet is one of the most popular datasets on the planet. Turns out, a significant fraction of its images are mislabeled. In the limit case the model would have to fit towards wrong answers to get higher than a certain percentage. The answer is “it works because ML wants to work.” It’s surprising how far you can get with something flawed. It’s also why such huge breakthroughs are possible by noting flaws others hav…
I do these sort of breakthroughs at home all the time! My wife would say the computer is doing something strange, and instead of just randomly clicking around, I read the error messages slowly and out loud, then follow what they say. Anyone can do this, yet it seems like a magical ability every time you employ it to help people.
Why don't they ask their premier model to generate a bench for them? Jokes aside, a benchmark I look forward to is ARC-AGI-3. I tried out their human simulation, and it feels very reasoning heavy. Leaderboard: https://arcprize.org/leaderboard (Most premier models don't even pass 5 percent.)
Issue with these benchmark also is that they measure a model you are unlikely going to be routed to. My experience with Anthropic is that despite using Opus 4.6 and 4.7, most of the time the performance is matching low B parameter Qwen. I think there should be a way to verify what model is actually being used to process prompts - that should be independently verified. At the moment it is so bad, you have to ask verif…
Why don't they ask their premier model to generate a bench for them? Jokes aside, a benchmark I look forward to is ARC-AGI-3. I tried out their human simulation, and it feels very reasoning heavy. Leaderboard: https://arcprize.org/leaderboard (Most premier models don't even pass 5 percent.)
Its pretty clear that any benchmark that comes out will be outdated and exist within the training data with short measure. There will always be an incentive to optimize specifically for these benchmarks even if just for marketing material. Sure there is a training cutoff, but its usually only 3-6 months off of the public release dates. The problem with coding benchmarks then becomes creating novel benchmarks that are…
Earlier quoted context omitted.
Imagenet is one of the most popular datasets on the planet. Turns out, a significant fraction of its images are mislabeled. In the limit case the model would have to fit towards wrong answers to get higher than a certain percentage. The answer is “it works because ML wants to work.” It’s surprising how far you can get with something flawed. It’s also why such huge breakthroughs are possible by noting flaws others hav…
> It’s also why such huge breakthroughs are possible by noting flaws others haven’t. I do these sort of breakthroughs at home all the time! My wife would say the computer is doing something strange, and instead of just randomly clicking around, I read the error messages slowly and out loud, then follow what they say. Anyone can do this, yet it seems like a magical ability every time you employ it to help people.
Its pretty clear that any benchmark that comes out will be outdated and exist within the training data with short measure. There will always be an incentive to optimize specifically for these benchmarks even if just for marketing material. Sure there is a training cutoff, but its usually only 3-6 months off of the public release dates. The problem with coding benchmarks then becomes creating novel benchmarks that are…
Its pretty clear that any benchmark that comes out will be outdated and exist within the training data with short measure. There will always be an incentive to optimize specifically for these benchmarks even if just for marketing material. Sure there is a training cutoff, but its usually only 3-6 months off of the public release dates. The problem with coding benchmarks then becomes creating novel benchmarks that are…
This is why I made Zork bench. Zork, the text adventure game, is in the training data for LLMs. It’s also deterministic. Therefore it should be easy for an LLM to play and complete. Yet they don’t. Understanding why is the goal of Zork bench. https://github.com/mnky9800n/zork-bench
The LLMs I have tested have terrible world models and intuitions for how actions change the environment. They're also not great at discerning and pursuing the right goals. They're like an infinitely patient five-year old with amazing vocabulary.
[1]: https://entropicthoughts.com/updated-llm-benchmark
(more descriptions available in earlier evaluations referenced from there)