AI Paper Atlas

Research map / Video / Video-language understanding

NarrativeTrack: Evaluating Entity-Centric Reasoning for Narrative Understanding

Hyeonjeong Ha, Jinjin Ge, Bo Feng, Kaixin Ma, Gargi Chakraborty

ECCV 2026 · Video › Video-language understanding · approach: Long video understanding · 4 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
ensemble Detectron2 and OWLv2 detections via spatial consensus (ablation)entity detection pipeline · AVA recallrecall: 0.780 / 0.801 -> 0.848Improved
scale open-source model size within a familyInternVL3 8B to 38B · NarrativeTrack avgaccuracy (%): 48.81 -> 55.47Improved
scale Video-LLaMA2 from 7B to 72BVideo-LLaMA2 · NarrativeTrack avgaccuracy (%): 43.04 -> 49.70Improved
scale LLaVA-NeXT-Video from 7B to 34BLLaVA-NeXT-Video · NarrativeTrack avgaccuracy (%): 42.94 -> 39.36Hurt
increase GPT-4o frames from 20 to 128 (ablation)GPT-4o · NarrativeTrack avgaccuracy (%): 70.97 -> 72.27Improved
increase frame count beyond 20 for open-source models (ablation)open-source MLLMs · NarrativeTrack avgaccuracy (%)Hurt
remove visual input (ablation)GPT-4o · NarrativeTrack avgaccuracy (%): -30.52%Hurt
reverse video frames (ablation)GPT-4o · entity change orderingaccuracy (%): 51.2 -> 6.1Hurt
backward instead of forward temporal reasoningMLLMs · NarrativeTrackaccuracy (%): gap of 20.65% (general), 9.96% (video), 17.55% (proprietary)Hurt
synthetic instead of real distractorsvideo-specialized MLLMs · NarrativeTrackaccuracy (%): up to 9.69% dropHurt
synthetic instead of real distractorsgeneral-purpose MLLMs · NarrativeTrackaccuracy (%): +0.44% on syntheticNo clear effect

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