Research
Researchers propose CodeCap-Bench to test whether code can serve as video captions
The paper compares code captions with natural language captions on video question answering and reconstruction, finding code captions score higher. It introduces CodeCap-Bench with 323 videos, where leading proprietary MLLMs perform reasonably and open-source models lag substantially.
Every outlet that ran it 1 article
Hugging Face PapersResearchLead