Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

MMToM-QA

Benchmarks

Task NameDataset NameSOTA ResultTrend
Theory of Mind reasoningMMToM-QA
Overall Accuracy98.5
44
Mental State InferenceMMToM-QA human 1.0 (test)
Sub-score 1.1100
20
Theory of Mind reasoningMMToM-QA Text-only
Belief Inference 1.11
17
Theory of Mind reasoningMMToM-QA Multimodal
Belief Inference 1.195.8
14
Theory of Mind reasoningMMToM-QA Video-only
Belief Inference 1.169.1
13
Showing 5 of 5 rows