🎯 PDB Leaderboard
Precision-aware evaluation of LLM debugging capabilities
PDB-Single Leaderboard
| # | Model | Precision â–¼ | Recall | Pass@1 |
|---|
PDB-Wild Leaderboard
| # | Model | Precision â–¼ | Recall | Pass@1 |
|---|
Leaderboard Submission
To submit a new model to the PDB leaderboard, please send an email to
miaosenc@uchicago.edu and
wangzhu@berkeley.edu with the subject line
PDB submission: <new model name>, <set name>, where
<set name> is one of the evaluation sets (e.g.
PDB-Single or PDB-Wild), and attach both
the generated result file and the score file produced by our evaluation pipeline.
We will verify the submission against our evaluation protocol and update the
leaderboard once the results are confirmed.