AI Tools & Models HOTFeaturedAll
arXiv cs.AIScore60

Vision-language models lag human performance on physical dynamics and intent reasoning

Summary

The study finds vision-language models still lag humans clearly in understanding physical interactions and inferring intent in open scenes, leaving spatial intelligence a weak spot.

Why this matters

Covers model · Relevant to this topic

Open original ↗

Source: arXiv cs.AI · arxiv.org

Topics

Related items