AI Paper Atlas

Research map / Video / Video-language understanding

VRR-QA: Visual Relational Reasoning in Videos Beyond Explicit Cues

Sirnam Swetha, Rohit Gupta, Parth Parag Kulkarni, David G Shatwell, Jeffrey A Chan Santiago, Nyle Siddiqui, Joseph Fioresi, Mubarak Shah

CVPR 2026 · Video › Video-language understanding · approach: Video question answering

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 a structured spatial and narrative reasoning prompt (ablation)GPT-4.1 Full · VRR-QAaccuracy (avg, %): 54.30 -> 58.20Improved
add a structured spatial and narrative reasoning prompt (ablation)GPT-4.1 Mini · VRR-QAaccuracy (avg, %): 40.30 -> 45.07Improved
increase frames from 8 to 16 (ablation)video LLMs · VRR-QAaccuracyImproved
increase frames from 16 to 32 (ablation)video LLMs · VRR-QAaccuracy: LLaVA-NeXT-Video 33.9 -> 32.56No clear effect
increase frames from 4 to 32 for inferred counting (ablation)Qwen2.5-VL-32B · VRR-QA Inferred Countingaccuracy (%): 27.91 -> 41.86Improved
increase frames from 1 to 32 (ablation)Qwen2.5-VL-7B · VRR-QAaccuracy (avg, %): 38.18 -> 45.07Improved
scale Qwen2.5-VL from 3B to 32B (ablation)Qwen2.5-VL · VRR-QAaccuracy: approximately +2% accuracyImproved
scale GPT-4.1 from Mini to Full (ablation)GPT-4.1 · VRR-QAaccuracy (avg, %): 40.30 -> 54.30Improved
use reasoning model GPT O3 instead of GPT 4.1GPT family · VRR-QAaccuracy (avg, %): 54.3 -> 64.1Improved
enable thinking tokens (ablation)Claude 4.5 Sonnet · VRR-QAaccuracy (avg, %): 42.8 -> 48.6Improved
enable thinking tokens (ablation)GPT 5.2 · VRR-QAaccuracy (avg, %): 50.8 -> 56.8Improved
allow annotators to view the video multiple times for counting (ablation)non-expert human baseline · VRR-QA Inferred Countingaccuracy (%): 65.9 -> 82.8Improved

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