mlsanityData to 14 Oct 2026

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

arXiv 2603.05256 ↗First posted 5 Mar 2026cs.CVcited by 2

On the map: LLM reasoning → Multimodal reasoning

Abstract (arXiv)

Knowledge-Based Visual Question Answering (KB-VQA) requires models to answer questions about an image by integrating external knowledge, posing significant challenges due to noisy retrieval and the structured, encyclopedic nature of the knowledge base. These characteristics create a distributional gap from pretrained multimodal large language models (MLLMs), making effective reasoning and domain adaptation difficult in the post-training stage. In this work, we propose textitWiki-R1, a data-generation-based curriculum reinforcement learning framework that systematically incentivizes reasoning in MLLMs for KB-VQA. Wiki-R1 constructs a sequence of training distributions aligned with the model's evolving capability, bridging the gap from pretraining to the KB-VQA target distribution. We introduce textitcontrollable curriculum data generation, which manipulates the retriever to produce samples at desired difficulty levels, and a textitcurriculum sampling strategy that selects informative samples likely to yield non-zero advantages during RL updates. Sample difficulty is estimated using observed rewards and propagated to unobserved samples to guide learning. Experiments on two KB-VQA benchmarks, Encyclopedic VQA and InfoSeek, demonstrate that Wiki-R1 achieves new state-of-the-art results, improving accuracy from 35.5% to 37.1% on Encyclopedic VQA and from 40.1% to 44.1% on InfoSeek. The project page is available at https://artanic30.github.io/projectpages/WikiR1/.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Multimodal reasoning, one of 17 directions in LLM reasoning. Builds on 27 of its 63 references that are in the map. First posted on arXiv 5 Mar 2026.

Builds on

27 of 63 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025VL-Cogito: Progressive Curriculum Reinforcement Learning for Advanced Multimodal Reasoning30 Jul

Other references · 26

  1. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  2. 2023BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Modelscited by 6,137
  3. 2025Qwen2.5-VL Technical Reportcited by 5,587
  4. 2023Improved Baselines with Visual Instruction Tuningcited by 4,418
  5. 2023InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuningcited by 2,876
  6. 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,810
  7. 2025Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learningcited by 1,444
  8. 2023EVA-CLIP: Improved Training Techniques for CLIP at Scalecited by 798
  9. 2017Automated Curriculum Learning for Neural Networkscited by 257
  10. 2023Can Pre-trained Vision and Language Models Answer Visual Information-Seeking Questions?cited by 207
  11. 2019Paired Open-Ended Trailblazer (POET): Endlessly Generating Increasingly Complex and Diverse Learning Environments and Their Solutionscited by 192
  12. 2023UniIR: Training and Benchmarking Universal Multimodal Information Retrieverscited by 166
14 more
  1. 2021KAT: A Knowledge Augmented Transformer for Vision-and-Languagecited by 126
  2. 2025MMSearch-R1: Incentivizing LMMs to Searchcited by 124
  3. 2019Teacher algorithms for curriculum learning of Deep RL in continuously parameterized environmentscited by 119
  4. 2022Retrieval Augmented Visual Question Answering with Outside Knowledgecited by 94
  5. 2025Efficient Reinforcement Finetuning via Adaptive Curriculum Learningcited by 93
  6. 2024EchoSight: Advancing Visual-Language Models with Wiki Knowledgecited by 62
  7. 2025DUMP: Automated Distribution-Level Curriculum Learning for RL-based LLM Post-trainingcited by 46
  8. 2024RoRA-VLM: Robust Retrieval-Augmented Vision Language Modelscited by 23
  9. 2023A Simple Baseline for Knowledge-Based Visual Question Answeringcited by 14
  10. 2024Mining Fine-Grained Image-Text Alignment for Zero-Shot Captioning via Text-Only Trainingcited by 12
  11. 2025Curriculum Guided Reinforcement Learning for Efficient Multi Hop Retrieval Augmented Generationcited by 5
  12. 2026DA-DPO: Cost-efficient Difficulty-aware Preference Optimization for Reducing MLLM Hallucinationscited by 3
  13. 2025A Curriculum Learning Approach to Reinforcement Learning: Leveraging RAG for Multimodal Question Answeringcited by 2
  14. 2025DACL-RAG: Data Augmentation Strategy with Curriculum Learning for Retrieval-Augmented Generationcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026WikiCLIP: An Efficient Contrastive Baseline for Open-domain Visual Entity Recognitioncited by 1
  2. 2026UniHEAR: Unified Heterogeneous-Source Attentive Retrieval for Knowledge-Based Visual Question Answeringcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025ReAG: Reasoning-Augmented Generation for Knowledge-based Visual Question Answering
  2. 2025Knowledge-based Visual Question Answer with Multimodal Processing, Retrieval and Filtering
  3. 2026Enhancing Visual Question Answering with Multimodal LLMs via Chain-of-Question Guided Retrieval-Augmented Generation
  4. 2023Language Guided Visual Question Answering: Elevate Your Multimodal Language Model Using Knowledge-Enriched Prompts
  5. 2024GeReA: Question-Aware Prompt Captions for Knowledge-based Visual Question Answering
  6. 2025Progressive Multimodal Search and Reasoning for Knowledge-Intensive Visual Question Answering
  7. 2026MaS-VQA: A Mask-and-Select Framework for Knowledge-Based Visual Question Answering
  8. 2026GRACE: Grounded Reasoning via Adapter Composition and Evidence-Aware Calibration for Educational Visual Question Answering