AI Tools & Models HOTFeaturedAll
arXiv cs.AIScore60

V-JEPA-LM: Video-Language Models That Read Dynamics Beyond Static, Text-Aligned Representations

Summary

This video-language model reads dynamics — direction, speed, order and state changes — instead of relying on static, text-aligned representations.

Why this matters

Covers model · Relevant to this topic

Open original ↗

Source: arXiv cs.AI · arxiv.org

Topics

Related items