Theory of Mind reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
95.89Accuracy
45
Jun 9, 2026
98.5Overall Accuracy
44
May 28, 2026
95.5Accuracy
24
Feb 26, 2026
98.2Accuracy
18
Apr 30, 2026
99.4Accuracy
18
Apr 30, 2026
1Belief Inference 1.1
17
May 12, 2026
100Accuracy (0th-Order)
16
Jun 11, 2026
96.131st-Order Accuracy
16
Jun 11, 2026
95.8Belief Inference 1.1
14
May 12, 2026
69.1Belief Inference 1.1
13
May 12, 2026
87.8Accuracy
12
Feb 26, 2026
100Location Accuracy (Cg, First)
10
Jun 9, 2026
83.3Accuracy
8
Jun 9, 2026
100Confidence Score
7
Mar 4, 2026
1Confidence Score
7
Mar 4, 2026
80Total Accuracy
7
Feb 26, 2026
85.6Belief Inference Accuracy
6
May 12, 2026
85.3Belief Inference Acc
6
May 12, 2026
87.2Belief Inference Accuracy
6
May 12, 2026
86Belief Inference Accuracy
6
May 12, 2026
85.8Belief Inference Accuracy
6
May 12, 2026
87Belief Inference Accuracy
6
May 12, 2026
67.5BeliefQA Accuracy
3
Jun 5, 2026