Scores for tikv-18498, ray-15990, ray-30229, ignite3-21776 and ignite3-22597 use the corrected v1.1 labels (OrderLab/distdebug-bench#4). Config order follows exact Table 1 block coverage. paper/reproduce.py on this build matches the paper for 17/17 configurations.
Mirrors OrderLab/distdebug-bench#4. tikv-17830 and tikv-19220 are rerun on their corrected snapshots for 14 configurations (the o4-mini rows keep their original attempts); every score uses the v1.1 labels. configs.json lists Kimi K2.6 above MiMo v2.5 Pro to match Table 1, and harness versions are recorded only where the run recorded them. paper/reproduce.py on this revision reproduces the v1.1 reference exactly.
