Structured video reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
22.7Exact Match
6
Jun 15, 2026
99Exact Match Rate
6
Jun 15, 2026
99Exact Match (%)
6
Jun 15, 2026
100Exact Match Accuracy
6
Jun 15, 2026
99.2Exact Match Percentage
6
Jun 15, 2026
88.7Exact Match Rate
6
Jun 15, 2026
89Success Rate
6
Jun 15, 2026
77Exact Match Rate
6
Jun 15, 2026
86.5Exact Match (%)
6
Jun 15, 2026
79.7Exact Match
6
Jun 15, 2026
79.5Exact Match (%)
6
Jun 15, 2026
99Exact Match (%)
6
Jun 15, 2026
82.7Exact Match Rate
6
Jun 15, 2026
95.3Exact Match
6
Jun 15, 2026
100Exact Match Rate
6
Jun 15, 2026
100Exact Match Percentage
6
Jun 15, 2026
97.7Exact Match (%)
6
Jun 15, 2026
81Exact Match Rate
6
Jun 15, 2026
99.5Exact Match Rate
6
Jun 15, 2026
100Exact Match Accuracy
6
Jun 15, 2026
90.3Exact Match Percentage
6
Jun 15, 2026