Pull requests: open-compass/opencompass
Pull requests list
[Fix] Support PromptBench package for prompt attack
#2555
opened Jul 16, 2026 by
ssiq
Collaborator
Loading…
fix: use last-match in generic LLM judge postprocess (verdict-injection defense)
#2534
opened Jul 14, 2026 by
AUTHENSOR
Loading…
3 tasks done
ci: pin jlumbroso/free-disk-space to a full commit SHA
#2533
opened Jul 13, 2026 by
kobihikri
Loading…
Fix wikitext ppl evaluation without references
#2532
opened Jul 13, 2026 by
ssiq
Collaborator
Loading…
[Fix] Handle tokenizers with read-only pad_token_id property
#2526
opened Jul 11, 2026 by
feiiiiii5
Loading…
4 of 5 tasks
feat(dataset): add CHHallu-Src v1 — Chinese history hallucination & source-attribution benchmark
#2525
opened Jul 11, 2026 by
lizhuojunx86
Loading…
fix(MedCalc_Bench): guard calid-69 ground_truth regex match (AttributeError)
#2523
opened Jul 9, 2026 by
WatchTree-19
Loading…
fix(evaluator): guard IndexError on truncated judge choice (JudgeEvaluator/RMBEvaluator)
#2522
opened Jul 9, 2026 by
WatchTree-19
Loading…
Add Helium Market Resolution benchmark
#2507
opened Jul 3, 2026 by
connerlambden
Loading…
1 of 2 tasks
[Fix] Extract the final answer in GPQA simple-eval predictions
#2496
opened Jun 27, 2026 by
Hibbert133
Contributor
Loading…
4 of 6 tasks
Add optional juryeval integration for LLM-as-Judge metrics
#2465
opened May 31, 2026 by
py-ai-dev
Loading…
[Fix] Combine split eval results in default summarizer
#2451
opened May 15, 2026 by
yhzhu99
Contributor
Loading…
ProTip!
Find all pull requests that aren't related to any open issues with -linked:issue.