AI Paper Atlas

Research map / Video / Video-language understanding

Keep It Simple: Multi-Key Episodic Memory Retrieval for Ultra-Long Video Understanding

Yeeun Choi, Youngbeom Yoo, Joon-Young Lee, Hyolim Kang, Seon Joo Kim

ECCV 2026 · Video › Video-language understanding · approach: Long video understanding · 2 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 neighbor filtering (ablation)MERIT (Qwen3-VL-8B) · EgoLifeQAaverage accuracy (%): 48.0 -> 54.2Improved
add neighbor filtering (ablation)MERIT (GPT-5) · EgoLifeQAaverage accuracy (%): 70.0 -> 71.2Improved
add neighbor filtering to oracle clip (ablation)Target Time Oracle (Qwen3-VL-8B) · EgoLifeQAaverage accuracy (%): 64.0 -> 65.8Improved
add neighbor filtering (ablation)MERIT (GPT-5) · EgoLifeQAhit rate: 0.40 -> 0.56Improved
use all four keys instead of the summary key alone (ablation)MERIT (Qwen3-VL-8B) · EgoLifeQAaverage accuracy (%): 49.4 -> 54.2Improved
use summary key instead of event key as single key (ablation)MERIT (Qwen3-VL-8B) · EgoLifeQAaccuracy (%): 44.6 -> 49.4Improved
add multi-key representation to flat retrieval (ablation)flat retrieval (Qwen3-VL-8B) · EgoLifeQAaverage accuracy (%): 46.6 -> 48.0Improved
add neighbor filtering to flat retrieval (ablation)flat retrieval (Qwen3-VL-8B) · EgoLifeQAaverage accuracy (%): 46.6 -> 52.8Improved
add both multi-key and neighbor filtering to flat retrieval (ablation)flat retrieval (Qwen3-VL-8B) · EgoLifeQAaverage accuracy (%): 46.6 -> 54.2Improved
retrieve top 10 clips instead of 5 (ablation)MERIT (Qwen3-VL-8B) · EgoLifeQAaverage accuracy (%): 54.2 -> 52.2Hurt
retrieve top 10 clips instead of 5 (ablation)MERIT (GPT-5) · EgoLifeQAaverage accuracy (%): 67.4 -> 71.2Improved
retrieve top 10 clips instead of 5 (ablation)MERIT (Gemini 2.5 Pro) · EgoLifeQAaverage accuracy (%): 61.4 -> 64.2Improved

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