AI Paper Atlas

Research map / Vision-language and multimodal / Visual question answering

Wiki-R1: Incentivizing Multimodal Reasoning for Knowledge-based VQA via Data and Sampling Curriculum

Shan Ning, Longtian Qiu, Xuming He

ICLR 2026 · Vision-language and multimodal › Visual question answering · approach: Universal VQA models · 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
Use DAPO RL instead of SFT (ablation)Qwen-2.5-VL 3B · Encyclopedic VQA / InfoSeekoverall accuracy: EVQA 25.1 -> 31.4; InfoSeek 29.5 -> 41.5Improved
Add data curriculum generation to DAPO (ablation)DAPO on Qwen-2.5-VL 3B · Encyclopedic VQA / InfoSeekoverall accuracy: EVQA 31.4 -> 34.5; InfoSeek 41.5 -> 43.0Improved
Add curriculum sampling without observation propagation (ablation)DAPO + data curriculum · Encyclopedic VQA / InfoSeekoverall accuracy: EVQA 34.5 -> 32.1; InfoSeek 43.0 -> 40.0Hurt
Add observation propagation to curriculum sampling (ablation)DAPO + data + sampling curriculum · Encyclopedic VQA / InfoSeekoverall accuracy: EVQA 32.1 -> 35.9; InfoSeek 40.0 -> 42.2Improved
Full Wiki-R1 vs data curriculum alone on InfoSeek (ablation)DAPO on Qwen-2.5-VL 3B · InfoSeek overallaccuracy: 43.0 -> 42.2 (EVQA 34.5 -> 35.9)Mixed
Use Sentence Transformer instead of TF-IDF for label-propagation similarity (ablation)Wiki-R1 3B · Encyclopedic VQA / InfoSeekoverall accuracy: EVQA 35.9 -> 36.3; InfoSeek 42.2 -> 43.0Improved
Use Wiki-R1 7B instead of best ReflectiVA retrieval modeKB-VQA · Encyclopedic VQA (all) and InfoSeek (all)accuracy: EVQA 35.5 -> 37.1; InfoSeek 40.1 -> 44.1Improved
Provide oracle Wikipedia entity: Wiki-R1 7B vs ReflectiVA on EVQA single-hopWiki-R1 7B · EVQA single-hop (oracle entity)accuracy: 75.2 -> 69.2Hurt

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