Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
37Accuracy
15
Apr 7, 2026
70.2Accuracy
15
Apr 7, 2026
67.59Accuracy
15
Mar 25, 2026
61.2Correct Rate
15
Mar 18, 2026
47OpQA Score
15
Mar 5, 2026
84GPQA Score
15
Mar 4, 2026
115.78Context Influence
15
Feb 27, 2026
0.131Mean Per-Step Regret
15
Feb 26, 2026
0.155Mean Per-Step Regret
15
Feb 26, 2026
0.135Mean per-step regret
15
Feb 26, 2026
0.157Mean Per-Step Regret
15
Feb 26, 2026
0.188Mean Per-Step Regret
15
Feb 26, 2026
0.185Mean Per-Step Regret
15
Feb 26, 2026
0.096Mean Regret
15
Feb 26, 2026
0.095Mean Per-Step Regret
15
Feb 26, 2026
70Average Accuracy
15
Feb 26, 2026
73.4Accuracy (Benign)
15
Feb 26, 2026
79.9Robustness Score (LLM-J)
15
Feb 26, 2026
79.3Rob. LLM-J Score
15
Feb 26, 2026
71Robustness Accuracy
15
Feb 26, 2026
66.4Robustness Accuracy
15
Feb 26, 2026
65.8Robustness Accuracy
15
Feb 26, 2026
64.2Robustness Accuracy
15
Feb 26, 2026
76Robustness Accuracy
15
Feb 26, 2026
0.7Robustness Accuracy
15
Feb 26, 2026