AI Paper Atlas

Research map / Video / Video-language understanding

APVR: Hour-Level Long Video Understanding with Adaptive Pivot Visual Information Retrieval

Hong Gao, Yiming Bao, Xuezhen Tu, Bin Zhong, Linan Yue, Min-Ling Zhang

AAAI 2026 · Video › Video-language understanding · approach: Video LLMs with fine-grained understanding · 6 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 expanded semantic information to the query (ablation)APVR Qwen2-VL-7B · LongVideoBenchAvg accuracy: 59.1 -> 60.9Improved
add expanded semantic information to the query (ablation)APVR Qwen2.5-VL-7B · LongVideoBenchAvg accuracy: 63.2 -> 64.5Improved
add expanded semantic information to the query (ablation)APVR VideoLLaMA3-7B · LongVideoBenchAvg accuracy: 62.3 -> 63.5Improved
increase pivot frames K from 256 to 1024 (ablation)APVR · LongVideoBenchAvg accuracy: 63.1 -> 64.9Improved
increase pivot frames K from 32 to 1024 (ablation)APVR · LongVideoBenchAvg accuracy: 54.3 -> 64.9Improved
balance CLIP and Grounding-DINO weight (lambda 0.5 vs 0.1) (ablation)APVR · LongVideoBenchAvg accuracy: 64.2 -> 64.9Improved
use 3 iterations with stride 4 instead of 2 iterations (ablation)APVR · LongVideoBenchAvg accuracy: 63.9 -> 64.9Improved
use 5 iterations instead of 3 (ablation)APVR · LongVideoBenchAvg accuracy: 64.9 -> 64.5No clear effect
add Pivot Frame Retrieval and Pivot Token Retrieval (ablation)Qwen2.5-VL-7B · LongVideoBenchaccuracy: 59.5 -> 64.9Improved
add APVR to Qwen2-VL-7BQwen2-VL-7B · MLVU devaccuracy: 66.9 -> 73.4Improved
add APVR to VideoLLaMA3-7BVideoLLaMA3-7B · VideoMME overallaccuracy: 66.2 -> 68.1Improved

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