FlowBench: Counterfactual Imagination Training for Spatial Intelligence
arXiv:2611.03937v1 Announce Type: new Abstract: Humans imagine scene changes under hypothetical motions, an ability central to spatial reasoning and robotic planning. However, this capability remains underexplored in vision-language models and world models. We formalize this capability as Counterfa…
Read the full story at arXiv cs.CV ↗
Timeline · 1 report
- 2026-11-04 05:00 · arXiv cs.CV
FlowBench: Counterfactual Imagination Training for Spatial Intelligence