block	column_1	column_2	column_3	column_4	column_5
1	Benchmark	Primary target	Input or setting	Dynamic change	Intent/ToM
1	ToMBench	ToM and social cognition	Text, bilingual MCQ	No	Direct
1	MMToM-QA	Goals, beliefs and plans	Multimodal household activity	Partial	Direct
1	EgoToM	Egocentric ToM	Ego4D video	Partial	Direct
1	MuMA-ToM	Multi-agent ToM	Household video and text	Partial	Direct
1	SpatialVLM	Metric and 3D space	Images and generated 3D VQA	No	No
1	VSI-Bench	Video spatial understanding	Real video and navigation QA	Direct	No
1	Spatial-DISE	Spatial-reasoning taxonomy	Real and synthetic VQA	Direct	No
1	EscherVerse	Goal-conditioned spatial change	11,328 open-world videos	Direct	Direct
2	Benchmark	Goal-conditioned change	Reference frame	Human comparison	Training resource
2	ToMBench	No	No	Yes	No
2	MMToM-QA	Partial	No	Yes	No
2	EgoToM	Partial	Partial	Yes	No
2	MuMA-ToM	Partial	No	Yes	No
2	SpatialVLM	No	Partial	No	Yes
2	VSI-Bench	No	Direct	Yes	No
2	Spatial-DISE	No	Direct	Yes	Yes
2	EscherVerse	Direct	Direct	Yes	Yes
