AI Paper Atlas

Research map / Video / Video-language understanding

LVBench: An Extreme Long Video Understanding Benchmark

Weihan Wang, Zehai He, Wenyi Hong, Yean Cheng, Xiaohan Zhang, Ji Qi, Ming Ding, Xiaotao Gu, Shiyu Huang, Bin Xu, Yuxiao Dong, Jie Tang

ICCV 2025 · Video › Video-language understanding · approach: Video LLMs with fine-grained understanding · 278 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
increase input frames from 50 to dense 1 FPS (ablation)Seed1.5-VL, Gemini-2.5-Pro, Qwen2.5-VL-72B · LVBenchaccuracyImproved
increase input frames from 1 to 8 (ablation)Seed1.5-VL, Gemini-2.5-Pro, Qwen2.5-VL-72B · LVBenchaccuracyImproved
give 0 frames (text-only input) (ablation)Seed1.5-VL, Gemini-2.5-Pro, Qwen2.5-VL-72B · LVBenchaccuracyHurt
use native long-video support instead of sampled frames from short-video modelsvideo MLLMs · LVBenchoverall accuracy: 22.5 (MovieChat) vs 45.3 (VideoLLaMA3-7B, non-native)Mixed
filter questions answerable by text-only LLMs (GLM-4 and GPT-4 both correct)LVBench construction · LVBenchdataset sizeImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy