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

VSR

Benchmarks

Task NameDataset NameSOTA ResultTrend
Visual Spatial ReasoningVSR
Accuracy88.87
59
Spatial ReasoningVSR
LLM-Judge Accuracy92.61
28
Visual Question AnsweringVSR
Top-1 Accuracy73.77
26
Spatial Relationship UnderstandingVSR
Overall Accuracy73.9
17
Relational ReasoningVSR
Accuracy85.7
16
2D Spatial ReasoningVSR
Accuracy75.6
10
Spatial ReasoningVSR (ood)
Accuracy84.8
10
Spatial ReasoningVSR (Visual Spatial Reasoning)
Binary Robust Acc75.8
9
Spatial Und. (Mono.)VSR (test)
Accuracy81.05
9
Directional attributionVSR (n=240)
DAE96.8
8
Visual Spatial ReasoningVSR ZOOM-Hard
GPT Accuracy55.98
6
Visual Spatial ReasoningVSR ZOOM-Medium
GPT Accuracy67.63
6
Visual Spatial ReasoningVSR (ZOOM-Easy)
GPT Accuracy73.09
6
Confidence estimationVSR (test)
AUROC67.4
6
Spatial ReasoningVSR zero-shot (test)
Accuracy (zero-shot)63.67
6
GeneralVSR
Score80.6
3
Showing 16 of 16 rows