Temporal Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
65.06F1 Score
45
Apr 15, 2026
74.4Accuracy
33
Apr 10, 2026
82.2EM
24
Feb 26, 2026
93.7EM
24
Feb 26, 2026
0.851EM
20
Feb 26, 2026
80.8EM
20
Feb 26, 2026
78.94Positive Score
17
Feb 26, 2026
63.77Positive Accuracy
17
Feb 26, 2026
77.45Positive Score
17
Feb 26, 2026
13.41Top-1 Accuracy
15
May 19, 2026
89.53Accuracy
14
Apr 7, 2026
81.9Visual Trace
13
Feb 26, 2026
88.9Location Accuracy
13
Feb 26, 2026
67.84Accuracy
9
Feb 26, 2026
60.3Exact Match (EM)
9
Feb 26, 2026
100Performance Score
8
Mar 10, 2026
15.03F1 Score
7
Feb 27, 2026
12.69F1 Score
7
Feb 27, 2026
0.742LLM Score
7
Feb 26, 2026
53.1Accuracy
6
Feb 27, 2026
40.2F1 Score
6
Feb 26, 2026
62Test Accuracy
4
Feb 26, 2026
56.8EM
4
Feb 26, 2026
53.4Accuracy
3
Jun 12, 2026
93Accuracy
3
Jun 2, 2026