mlsanityData to 14 Oct 2026

Self-Critical Reasoning for Robust Visual Question Answering

arXiv 1905.09998 ↗First posted 24 May 2019cs.CV cs.CLcited by 61

Not on the map. Why?

Abstract (arXiv)

Visual Question Answering (VQA) deep-learning systems tend to capture superficial statistical correlations in the training data because of strong language priors and fail to generalize to test data with a significantly different question-answer (QA) distribution. To address this issue, we introduce a self-critical training objective that ensures that visual explanations of correct answers match the most influential image regions more than other competitive answer candidates. The influential regions are either determined from human visual/textual explanations or automatically from just significant words in the question and answer. We evaluate our approach on the VQA generalization task using the VQA-CP dataset, achieving a new state-of-the-art i.e., 49.5% using textual explanations and 48.5% using automatically annotated regions.

Key facts. Cited by 61 papers in the map (data through 14 Oct 2026). First posted on arXiv 24 May 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 17; 2022: 15; 2023: 13; 2024: 10; 2025: 4; 2026: 2.

61 papers.

Sort:

By area:

  1. 2021A Review on Explainability in Multimodal Deep Neural Netscited by 46
  2. 2022SwapMix: Diagnosing and Regularizing the Over-Reliance on Visual Context in Visual Question Answeringcited by 38
  3. 2022Going Beyond XAI: A Systematic Survey for Explanation-Guided Learningcited by 36
  4. 2021Greedy Gradient Ensemble for Robust Visual Question Answeringcited by 34
  5. 2022Language Prior Is Not the Only Shortcut: A Benchmark for Shortcut Learning in VQAcited by 30
  6. 2021Coarse-to-Fine Reasoning for Visual Question Answeringcited by 25
  7. 2021Introspective Distillation for Robust Question Answeringcited by 25
  8. 2021Separating Skills and Concepts for Novel Visual Question Answeringcited by 24
  9. 2023Robust Visual Question Answering: Datasets, Methods, and Future Challengescited by 23
  10. 2021Counterfactual Samples Synthesizing and Training for Robust Visual Question Answeringcited by 22
  11. 2023Debiased Fine-Tuning for Vision-language Models by Prompt Regularizationcited by 21
  12. 2021Multimodal Integration of Human-Like Attention in Visual Question Answeringcited by 20
38 more
  1. 2022On Modality Bias Recognition and Reductioncited by 20
  2. 2021AdaVQA: Overcoming Language Priors with Adapted Margin Cosine Losscited by 18
  3. 2022Visual Perturbation-aware Collaborative Learning for Overcoming the Language Prior Problemcited by 18
  4. 2024Look, Listen, and Answer: Overcoming Biases for Audio-Visual Question Answeringcited by 18
  5. 2021LPF: A Language-Prior Feedback Objective Function for De-biased Visual Question Answeringcited by 17
  6. 2022Overcoming Language Priors in Visual Question Answering via Distinguishing Superficially Similar Instancescited by 16
  7. 2024GazeXplain: Learning to Predict Natural Language Explanations of Visual Scanpathscited by 15
  8. 2021Discovering the Unknown Knowns: Turning Implicit Knowledge in the Dataset into Explicit Training Examples for Visual Question Answeringcited by 14
  9. 2022REX: Reasoning-aware and Grounded Explanationcited by 12
  10. 2022VisFIS: Visual Feature Importance Supervision with Right-for-the-Right-Reason Objectivescited by 12
  11. 2022Generative Bias for Robust Visual Question Answeringcited by 12
  12. 2024Unveiling and Mitigating Bias in Audio Visual Segmentationcited by 11
  13. 2021X-GGM: Graph Generative Modeling for Out-of-Distribution Generalization in Visual Question Answeringcited by 10
  14. 2023S3C: Semi-Supervised VQA Natural Language Explanation via Self-Critical Learningcited by 10
  15. 2021Answer Questions with Right Image Regions: A Visual Attention Regularization Approachcited by 9
  16. 2022Towards Robust Visual Question Answering: Making the Most of Biased Samples via Contrastive Learningcited by 9
  17. 2024Uncovering the Full Potential of Visual Grounding Methods in VQAcited by 9
  18. 2023Measuring Faithful and Plausible Visual Grounding in VQAcited by 8
  19. 2024A Comprehensive Survey on Visual Question Answering Datasets and Algorithmscited by 8
  20. 2021Cross-Modal Generative Augmentation for Visual Question Answeringcited by 7
  21. 2021General Greedy De-bias Learningcited by 7
  22. 2022Guiding Visual Question Answering with Attention Priorscited by 7
  23. 2022Entity-Focused Dense Passage Retrieval for Outside-Knowledge Visual Question Answeringcited by 7
  24. 2021Language bias in Visual Question Answering: A Survey and Taxonomycited by 6
  25. 2023Towards More Faithful Natural Language Explanation Using Multi-Level Contrastive Learning in VQAcited by 5
  26. 2024On the Role of Visual Grounding in VQAcited by 5
  27. 2023Divide and Conquer: Answering Questions with Object Factorization and Compositional Reasoningcited by 4
  28. 2023From Wrong To Right: A Recursive Approach Towards Vision-Language Explanationcited by 4
  29. 2023Debiasing Multimodal Models via Causal Information Minimizationcited by 4
  30. 2021Explaining the Road Not Takencited by 3
  31. 2022Visually Grounded VQA by Lattice-based Retrievalcited by 3
  32. 2023An Empirical Study on the Language Modal in Visual Question Answeringcited by 3
  33. 2024Eliminating the Language Bias for Visual Question Answering with fine-grained Causal Interventioncited by 3
  34. 2023Object Attribute Matters in Visual Question Answeringcited by 2
  35. 2022Attention in Reasoning: Dataset, Analysis, and Modelingcited by 1
  36. 2023Unveiling Cross Modality Bias in Visual Question Answering: A Causal View with Possible Worlds VQAcited by 1
  37. 2024The Master-Slave Encoder Model for Improving Patent Text Summarization: A New Approach to Combining Specifications and Claimscited by 1
  38. 2025Overcoming Language Priors for Visual Question Answering Based on Knowledge Distillationcited by 1

50 most cited of 61 All 61 (JSON)