AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

V-REX: Benchmarking Exploratory Visual Reasoning via Chain-of-Questions

Chenrui Fan, Yijun Liang, Shweta Bhardwaj, Kwesi Cobbina, Ming Li, Tianyi Zhou

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Chain-of-thought multimodal reasoning · 1 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
provide ground-truth CoQ answers/questions as hints (Planning mode)GPT-5 · V-REX final questionaccuracy: 77.1 -> 86.0Improved
provide ground-truth CoQ questions (Following mode)GPT-5 · V-REX final questionaccuracy: 77.1 -> 81.4Improved
provide CoQ hints (Planning mode)InternVL3-38B · V-REX final questionaccuracy: 67.4 -> 72.0Improved
provide CoQ hints (Following mode)InternVL3-38B · V-REX final questionaccuracy: 67.4 -> 72.8Improved
provide CoQ Planning hintsQwen2.5-VL-7B · V-REX final question (Deduction)accuracy: 60.8 -> 35.2Hurt
provide CoQ Planning hintsInternVL3.5-1B · V-REX final question (overall)accuracy: 43.3 -> 37.7Hurt
provide CoQ hints on Retrieval categoryVLMs · V-REX RetrievalaccuracyNo clear effect
increase model sizeVLMs on V-REX · Planning vs Following averageaccuracy: 35.1 -> 84.4Improved
increase candidate questions from 4 to 8GPT-4o · V-REX Planning (average)accuracy: 72.72 -> 41.06Hurt
increase candidate questions from 4 to 8InternVL3-38B · V-REX Planning (average)accuracy: 81.57 -> 58.08Hurt
increase candidate questions from 4 to 8Qwen3-VL-2B-IT · V-REX Planning (average)accuracy: 32.96 -> 14.25Hurt
remove image input (blindfold)GPT-5 · V-REX Navigationaccuracy: 77.55No 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