BedBench reveals weaknesses in vision-language 3D reasoning
Best vision-language models achieve only 7.2% on per-color counting in BedBench compared to humans at 93.3%.
Stories per day · 30 days
1 RSS
Relatedvision-language-models3act3d1bedbench1benchmark1gpt-5.61reinforcement-learning1rota-bench1spatial-exploration1spatial-reasoning1
5 stories, newest first · one card per story
Stories tagged #3d-reasoning
ACT3D uses reinforcement learning so VLMs explore 3D from one view
Average accuracy reached 53.2% on six 3D benchmarks, 22.0 points above base Qwen and above Gemini 3.6 Flash.
RoTA-Bench: Evaluating 3D Spatial Reasoning through Interactive Projection Alignment
RoTA-Bench shows vision-language models struggle to recover target 3D projections through successive rotations.
CoVeR: Coverage-Based Token Pruning for Multi-View 3D Reasoning in VLMs
CoVeR, a training-free coverage-based token pruner, preserves 93.5% of VLM 3D-reasoning performance using only about 8% of visual tokens.
SpatialBlock: Enhancing Spatial Intelligence in LVLMs via Synthetic Block-Stacking Problem
Researchers release SpatialBlock-15k, a synthetic block-stacking dataset that improves 3D spatial reasoning in large vision-language models without dense geometric annotations.