GitHub / dvlab-research / LSDBench
A benchmark that focuses on the sampling dilemma in long-video tasks. Through well-designed tasks, it evaluates the sampling efficiency of long-video VLMs.
JSON API: http://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dvlab-research%2FLSDBench
Stars: 3
Forks: 0
Open issues: 0
License: apache-2.0
Language: Python
Size: 2.57 MB
Dependencies parsed at: Pending
Created at: 24 days ago
Updated at: 22 days ago
Pushed at: 22 days ago
Last synced at: 22 days ago
Topics: benchmark, long-video-understanding, multimodal-large-language-models, reasoning-agent, sampling-strategies, video-understanding, video-understanding-dataset, vision-language-model