AI Paper Atlas

Research map / Video / Video-language understanding

EventMemAgent: Hierarchical Event-Centric Memory for Online Video Understanding with Adaptive Tool Use

Siwei Wen, Zhangcheng Wang, Xingjian Zhang, Lei Huang, wenjun wu

ECCV 2026 · Video › Video-language understanding · approach: Long video understanding · 15 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 event-centric hierarchical memory instead of fixed-length 30s memory (ablation)EventMemAgent · OVO-Benchoverall accuracy: 58.27 -> 60.75Improved
add event-centric hierarchical memory instead of fixed-length 30s memory (ablation)EventMemAgent · StreamingBenchoverall accuracy: 75.09 -> 77.00Improved
add object detection tool (ablation)EventMemAgent · OVO-Benchoverall accuracy: 57.80 -> 60.75Improved
add object detection tool (ablation)EventMemAgent · OVO-Bench Real-Time Visual Perceptionavg accuracy: 65.33 -> 68.29Improved
add OCR tool (ablation)EventMemAgent · OVO-Benchoverall accuracy: 58.98 -> 60.75Improved
add OCR tool (ablation)EventMemAgent · OVO-Bench OCR taskaccuracy: 64.50 -> 75.84Improved
add agentic RL (GRPO) to the untrained agent (ablation)EventMemAgent · OVO-Benchoverall accuracy: 56.16 -> 60.75Improved
add agentic RL (GRPO) to the untrained agent (ablation)EventMemAgent · StreamingBenchoverall accuracy: 68.44 -> 77.00Improved
add memory and tools without RL to base MLLM (ablation)Qwen3-VL-8B · OVO-Bench Backward Tracingavg accuracy: 46.18 -> 56.33Improved
add memory and tools without RL to base MLLM (ablation)Qwen3-VL-8B · OVO-Bench Real-Time Visual Perceptionavg accuracy: 65.52 -> 62.85Hurt
add memory and tools without RL to base MLLM (ablation)Qwen3-VL-8B · StreamingBenchoverall accuracy: 70.20 -> 68.44Hurt
add memory and tools without RL to base MLLM (ablation)Qwen3-VL-8B · OVO-Benchoverall accuracy: 55.81 -> 56.16No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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