AI Paper Atlas

Research map / Video / Video-language understanding

Map the Flow: Revealing Hidden Pathways of Information in VideoLLMs

Minji Kim, Taekyung Kim, Bohyung Han

ICLR 2026 · Video › Video-language understanding · approach: Long video understanding · 5 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 cross-frame attention in first 16 layers (ablation)LLaVA-NeXT-7B-Video-FT · TVBench Object Countaccuracy drop when removed: -60.8%Improved
add cross-frame attention in first 16 layers (ablation)LLaVA-NeXT-7B-Video-FT · TVBench Action Antonymaccuracy drop when removed: -24.1%Improved
add cross-frame attention in first 16 layers (ablation)LLaVA-NeXT-7B-Video-FT · TVBench Scene Transitionaccuracy drop when removed: -18.0%Improved
restrict attention to effective pathways instead of full causal attentionLLaVA-NeXT-7B-Video-FT · TVBenchaccuracy: 51.5 -> 51.2No clear effect
restrict attention to effective pathways instead of full causal attentionLLaVA-NeXT-7B-Video-FT · TOMATOaccuracy: 30.2 -> 29.2No clear effect
restrict attention to effective pathways instead of full causal attentionLLaVA-NeXT-13B-Video-FT · TVBenchaccuracy: 55.1 -> 54.6No clear effect
restrict attention to effective pathways instead of full causal attentionVideoLLaMA3-7B · TVBenchaccuracy: 55.2 -> 57.2Improved
restrict attention to effective pathways instead of full causal attentionMini-InternVL-4B-Video-FT · TOMATOaccuracy: 32.2 -> 31.2No clear effect
use identified effective pathways instead of random blocking of same edge proportion (ablation)LLaVA-NeXT-7B-Video-FT · TVBenchaccuracy: 40.1 -> 51.2Improved
use identified effective pathways instead of random blocking of same edge proportion (ablation)VideoLLaMA3-7B · TVBenchaccuracy: 22.2 -> 57.2Improved
use identified effective pathways instead of random blocking of same edge proportion (ablation)Mini-InternVL-4B-Video-FT · TOMATOaccuracy: 25.9 -> 31.2Improved
use identified effective pathways instead of random blocking of same edge proportion (ablation)LLaVA-NeXT-13B-Video-FT · TOMATOaccuracy: 23.8 -> 27.4Improved

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