V-JEPA-LM: Video-Language Models That Read Dynamics Beyond Static, Text-Aligned Representations
V-JEPA-LM argues that video models can encode motion without using it. Counterfactual training makes speed, direction, and state changes answerable.
ls ./papers | grep "#video-language models"
Trending AI research papers from Hugging Face, summarized in plain language: what each paper does, why it matters and where to read more.
1 paper · #video-language models
V-JEPA-LM argues that video models can encode motion without using it. Counterfactual training makes speed, direction, and state changes answerable.