mattymchen/codejudgebench-adversarial download history

mattymchen/codejudgebench-adversarial is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 53 times (18 in the last 7 days), and 381 times in total. It ranks #177,952 among datasets by monthly downloads.

CodeJudgeBench-Adversarial CodeJudgeBench is a benchmark aimed at evaluating LLM-based judges for coding related tasks — see LINK. The adversarial version of CodeJudgeBench makes judging harder by having an LLM apply semantics-preserving adversarial transformations (e.g., variable

Open mattymchen/codejudgebench-adversarial on Hugging Face

Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.