AI Paper Atlas

Research map / Video / Video-language understanding

LvBench: A Benchmark for Long-form Video Understanding with Versatile Multi-modal Question Answering

Hongjie Zhang, Lu Dong, Yi Liu, Yifei Huang, Limin Wang, Limin Wang, Yu Qiao

IJCV 2025 · Video › Video-language understanding · approach: Video question answering · 48 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
add dual pathway modeling (DPM) to SEVILASEVILA · LvBench test finetunedaccuracy: 35.0 -> 36.9Improved
add DPM to LongVALongVA 7B · LvBench testaccuracy: 44.4 -> 46.6Improved
add DPM to LLaVA-OVLLaVA-OV 7B · LvBench testaccuracy: 46.5 -> 48.3Improved
add keyframe pathway to long context pathway (ablation)DPM · LvBench testaccuracy: 36.0 -> 36.9Improved
add long context pathway to keyframe pathway (ablation)DPM · LvBench testaccuracy: 36.3 -> 36.9Improved
increase input frames from 128 to 512 (ablation)DPM · LvBench testaccuracy: 36.2 -> 36.9Improved
increase input frames from 10 to 384 (ablation)GPT-4o · LvBenchaccuracy: 52.8 -> 67.3Improved
increase input frames from 10 to 1000 (ablation)Gemini-1.5 Pro · LvBenchaccuracy: 57.2 -> 71.5Improved
use ChatGPT QA-based distractors instead of manual distractors (ablation)FrozenBiLM · LvBenchaccuracy: 31.5 -> 79.6Hurt
fine-tune on LvBench instead of zero-shotSEVILA · LvBenchaccuracy: 28.2 -> 35.0Improved
add subtitles and video to question for humans (ablation)human · LvBenchaccuracy: 40.2 -> 91.3Improved
scale LLaVA-OV from 7B to 72BLLaVA-OV · LvBench zero-shotaccuracy: 40.4 -> 49.4Improved

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