Comment by parasmadan
7 days ago
Why not just use an eval harness to prove this catches more real bugs? Benchmarks on actual bug classes would be far more convincing than comparing against /review.
7 days ago
Why not just use an eval harness to prove this catches more real bugs? Benchmarks on actual bug classes would be far more convincing than comparing against /review.
That’s probably more work than the entire repo itself. Would need to be something like SWE-bench with and without “adamsreview”.
You’re right though, but evals are actually fairly tricky to write and maintain.
That's a great idea. I had trouble finding anything like this, a benchmark made for (AI) code reviewers.
I had expected to find something like an eval harness available on GitHub, but couldn't find it.
Any suggestions? Or maybe we/I/someone should build something like this?
I suppose one challenge is that if it's going to be publicly available, it would also be easy to cheat, but still seems it would be useful if people agreed it's a good benchmark and could easily re-test tools themselves.
https://www.codereviewbench.com/
https://codereview.withmartian.com/
Based on how same models rank fluctuates week to week, all I can conclude is that no frontier models is statistically better than the other or it's too task dependent that the result cannot converge.
I will be trying these and will report back!
[flagged]