mlsanityData to 14 Oct 2026

Why Did the Chicken Cross the Road? Rephrasing and Analyzing Ambiguous Questions in VQA

arXiv 2211.07516 ↗First posted 14 Nov 2022cs.CLcited by 13

Not on the map. Why?

Abstract (arXiv)

Natural language is ambiguous. Resolving ambiguous questions is key to successfully answering them. Focusing on questions about images, we create a dataset of ambiguous examples. We annotate these, grouping answers by the underlying question they address and rephrasing the question for each group to reduce ambiguity. Our analysis reveals a linguistically-aligned ontology of reasons for ambiguity in visual questions. We then develop an English question-generation model which we demonstrate via automatic and human evaluation produces less ambiguous questions. We further show that the question generation objective we use allows the model to integrate answer group information without any direct supervision.

Key facts. Cited by 13 papers in the map (data through 14 Oct 2026). Builds on 17 of its 48 references that are in the map. First posted on arXiv 14 Nov 2022.

Builds on

17 of 48 references.

References · 17

  1. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  2. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,512
  3. 2016Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answeringcited by 2,516
  4. 2014CIDEr: Consensus-based Image Description Evaluationcited by 2,046
  5. 2021ViLT: Vision-and-Language Transformer Without Convolution or Region Supervisioncited by 1,230
  6. 2018VizWiz Grand Challenge: Answering Visual Questions from Blind Peoplecited by 858
  7. 2020AmbigQA: Answering Ambiguous Open-domain Questionscited by 390
  8. 2018Fast Lexically Constrained Decoding with Dynamic Beam Allocation for Neural Machine Translationcited by 182
  9. 2021SituatedQA: Incorporating Extra-Linguistic Contexts into QAcited by 145
  10. 2019Cycle-Consistency for Robust Visual Question Answeringcited by 105
  11. 2016Generating Natural Questions About an Imagecited by 56
  12. 2017Visual Question Generation as Dual Task of Visual Question Answeringcited by 41
5 more
  1. 2019Information Maximizing Visual Question Generationcited by 40
  2. 2019Why Does a Visual Question Have Different Answers?cited by 34
  3. 2017iVQA: Inverse Visual Question Answeringcited by 22
  4. 2021Guiding Visual Question Generationcited by 14
  5. 2020Rephrasing visual questions by specifying the entropy of the answer distributioncited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 5; 2024: 2; 2025: 4; 2026: 2.

13 papers.

Sort:

By area:

  1. 2023We're Afraid Language Models Aren't Modeling Ambiguitycited by 126
  2. 2023Zero and Few-shot Semantic Parsing with Ambiguous Inputscited by 16
  3. 2023VQA Therapy: Exploring Answer Differences by Visually Grounding Answerscited by 14
  4. 2023Rephrase, Augment, Reason: Visual Grounding of Questions for Vision-Language Modelscited by 14
  5. 2025Teaching Vision-Language Models to Ask: Resolving Ambiguity in Visual Questionscited by 11
  6. 2024RAcQUEt: Unveiling the Dangers of Overlooked Referential Ambiguity in Visual LLMscited by 8
  7. 2025Acknowledging Focus Ambiguity in Visual Questionscited by 8
  8. 2023That was the last straw, we need more: Are Translation Systems Sensitive to Disambiguating Context?cited by 3
  9. 2025Trick or Neat: Adversarial Ambiguity and Language Model Evaluationcited by 3
  10. 2025Language Models Identify Ambiguities and Exploit Loopholescited by 2
  11. 2024SimpsonsVQA: Enhancing Inquiry-Based Learning with a Tailored Datasetcited by 1
  12. 2026LaViSA: A Language and Vision Structural Ambiguity Benchmarkcited by 0
1 more
  1. 2026CRAG-MM-Diagnostics: Enabling Stage-Wise Analysis of Knowledge-Intensive VQAcited by 0