AI Paper Atlas

Research map / Video / Video-language understanding

Em-Garde: A Propose-Match Framework for Proactive Streaming Video Understanding

Yikai Zheng, Xin Ding, Yifan Yang, Shiqi Jiang, Hao Wu, Qianxi Zhang, Weijun Wang, Ting Cao, Yunxin Liu

ECCV 2026 · Video › Video-language understanding · approach: Long video understanding · 13 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 IGPP proposal parser (ablation)Em-Garde · OVO-Bench Forward Active RespondingF1: 24.0 -> 31.0Improved
add IGPP proposal parser (ablation)Em-Garde · StreamingBench Proactive Outputaccuracy: 28.8 -> 38.0Improved
add LPMM embedding matching (ablation)Em-Garde · OVO-Bench Forward Active RespondingF1: 14.6 -> 31.0Improved
add LPMM embedding matching (ablation)Em-Garde · StreamingBench Proactive Outputaccuracy: 23.2 -> 38.0Improved
propose-match decomposition versus naive sliding-window (ablation)Em-Garde · OVO-Bench Forward Active RespondingF1: 21.0 -> 31.0Improved
add RL stage after SFT (lambda=1) (ablation)IGPP · OVO-Bench Forward Active RespondingF1: 23.4 -> 31.0Improved
add false-positive penalty lambda=1 versus 0 (ablation)IGPP RL · OVO-Bench Forward Active RespondingF1: 27.7 -> 31.0Improved
raise lambda from 1 to 1.5 (ablation)IGPP RL · OVO-Bench Forward Active RespondingF1: 31.0 -> 19.3Hurt
add false-positive penalty lambda=0 to RL on StreamingBench PO (ablation)IGPP RL · StreamingBench Proactive Outputaccuracy: SFT 34.8 -> RL lambda=0 30.8Hurt
match training and test threshold at 0.04 versus 0.03 (ablation)Em-Garde · OVO-Bench Forward Active RespondingAvg F1: 26.59 -> 30.99Improved
raise streaming fps from 1 to 2 (ablation)Em-Garde · StreamingBench Proactive Outputaccuracy: 37.6 -> 38.0Improved
use Em-Garde instead of MMDuet-2 on ProactiveVideoQA WEBEm-Garde · ProactiveVideoQA WEBPAUC: 53.3 -> 44.3Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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