Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
1.041Rel. Perf vs Truncated ICL
5
Feb 26, 2026
97.85EM
5
Feb 26, 2026
6.2Training Time (hours)
5
Feb 26, 2026
32.26Accuracy
5
Feb 26, 2026
43.36Accuracy
5
Feb 26, 2026
26.58Accuracy
5
Feb 26, 2026
28.88Accuracy
5
Feb 26, 2026
60Exact Match (EM)
5
Feb 26, 2026
0.77LLM Judge Score
5
Feb 26, 2026
50LLM-as-a-Judge Score
5
Feb 26, 2026
0Indexing Time (s)
5
Feb 26, 2026
41.65F1 (Answer)
5
Feb 26, 2026
70.01AvgQA Score
5
Feb 26, 2026
70Match
5
Feb 26, 2026
91.9Match Score
5
Feb 26, 2026
51.9Match
5
Feb 26, 2026
67.7Match Score
5
Feb 26, 2026
94.9QA-Causal Score
5
Feb 26, 2026
65.81F1 Score
5
Feb 26, 2026
84.32F1 Score
5
Feb 26, 2026
93Accuracy
5
Feb 26, 2026
95Accuracy
5
Feb 26, 2026
77.72Accuracy
5
Feb 26, 2026
72.52Accuracy
5
Feb 26, 2026
71.34Accuracy
5
Feb 26, 2026