AI Paper Atlas

Research map / Video / Video-language understanding

Long-range Modeling and Processing of Multimodal Event Sequences

Jichu Li, Yilun Zhong, Zhiting Li, Feng Zhou, Quyu Kong

ICLR 2026 · Video › Video-language understanding · approach: Video LLMs with fine-grained understanding · 0 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 adaptive temporal-similarity sequence compression (ablation)MM-TPP · DanmakuTPPRMSE: 5.5551 -> 5.2987Improved
add adaptive temporal-similarity sequence compression (ablation)MM-TPP · DanmakuTPPACC (%): 25.87 -> 27.62Improved
use adaptive compression instead of random event drop p=25% (ablation)MM-TPP · DanmakuTPPRMSE: 5.3792 -> 5.2987Improved
use adaptive compression instead of random event drop p=50% (ablation)MM-TPP · DanmakuTPPACC (%): 24.79 -> 27.62Improved
add image input to text-only MM-TPP (ablation)MM-TPP · DanmakuTPPRMSE: 5.4654 -> 5.2987Improved
add image input to text-only MM-TPP (ablation)MM-TPP · DanmakuTPPACC (%): 23.64 -> 27.62Improved
add image input to text-only MM-TPP (ablation)MM-TPP · TAXI-PROACC (%): 76.70 -> 77.56Improved
scale backbone from Qwen2.5-VL-3B to 7B (ablation)MM-TPP · DanmakuTPPRMSE: 5.2987 -> 5.0533Improved
scale backbone from Qwen2.5-VL-3B to 7B (ablation)MM-TPP · DanmakuTPPACC (%): 27.62 -> 26.98Hurt
scale backbone from Qwen2.5-VL-3B to 7B (ablation)MM-TPP · TAXI-PROACC (%): 77.56 -> 76.16Hurt
increase max token limit from 2048 to 4096 (ablation)MM-TPP · DanmakuTPPRMSE and ACCImproved
use similarity threshold 0.2 instead of 0.05 or 0.5 (ablation)MM-TPP · DanmakuTPPRMSE and ACCImproved

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