AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Ego3S: Select, Strengthen, and Synchronize for Efficient Egocentric Reasoning

Shenshen Li, Kaiyuan Deng, RuoHuai Xie, Xing Xu, Heng Tao Shen, Yazhou Yao, Fumin Shen

ICML 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Chain-of-thought multimodal reasoning

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Ego3S proposes a select-strengthen-synchronize framework to make large vision-language models reason efficiently over egocentric video.

Problem

Egocentric reasoning in large vision-language models is harder than third-person understanding because decisive cues are localized, atypical and scenes are highly dynamic.

Approach

Frames egocentric reasoning as needing focus on localized decisive cues, and proposes a select, strengthen, and synchronize pipeline to achieve efficient egocentric reasoning; the abstract does not specify details.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add coarse counterfactual filter (CCF) to full-data training (ablation)Qwen2.5-VL-7B GRPO · QAEgo4Dpass@1: 63.80 -> 64.80Improved
add fine counterfactual filter (FCF) to full-data training (ablation)Qwen2.5-VL-7B GRPO · QAEgo4Dpass@1: 63.80 -> 64.60Improved
combine CCF and FCF (full CSM) (ablation)Qwen2.5-VL-7B GRPO · QAEgo4Dpass@1: 63.80 -> 65.80Improved
add interaction reward rint on top of CSM (ablation)CSM · QAEgo4Dpass@1: 65.80 -> 66.20Improved
add penalty term rp on top of CSM+rint (ablation)CSM+rint · EgoTextVQApass@1: 21.60 -> 22.00Improved
add Evolving Curriculum Recuration (ECR) (ablation)CSM+IRL · EgoTextVQApass@1: 22.00 -> 22.40Improved
replace full-data training with CSM selection (26.5% of data)GRPO · Egoplan2pass@1: 30.43 -> 33.76Improved
CSM selection instead of full data: training timeGRPO · Training costtime (h): 60.5 -> 29.7Improved
use CSM instead of LIMR data selectionQwen2.5-VL-7B · Egoplan2pass@1: 31.46 -> 33.76Improved
use IRL instead of GRPOQwen2.5-VL-7B · Egoplan2pass@1: 30.43 -> 35.42Improved
use IRL instead of DAPOQwen2.5-VL-7B · Egoplan2pass@1: 30.87 -> 35.42Improved
use RLOO instead of GRPOQwen2.5-VL-7B · EgoSchemapass@1: 47.60 -> 46.90Hurt

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