mlsanityData to 14 Oct 2026

Calibrated Ambiguity in Multimodal Language Models: Humans reach for cultural references, while models describe the picture

arXiv 2609.12575 ↗First posted 11 Sep 2026cs.CL cs.AI cs.HCcited by 0

On the map: Multimodal LLMs / vision-language models → VLM benchmarks and visual reasoning

Abstract (arXiv)

Ambiguity is often treated as a bug for AI systems to resolve---but in human communication and culture, ambiguity can also be a generative resource. From humour to politics to art, people express themselves in words and images that are open enough to invite different interpretations, yet constrained enough to be interpretable. We operationalise this notion of calibrated ambiguity with a task drawn from the parlour game Dixit. We compare differences in clues generated by human vs multimodal language models, based on a novel coding rubric for calibrated ambiguity, and find that models consistently exhibit ambiguity collapse (i.e., their outputs are over-specified, leaving no room for multiple legitimate interpretations). Unlike human clues, AI-generated clues also exhibit cultural flattening; they almost never make reference to culturally-situated knowledge, even when prompted to use allusion and figurative language.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in VLM benchmarks and visual reasoning, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 12 of its 83 references that are in the map. First posted on arXiv 11 Sep 2026.

Builds on

12 of 83 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025Evaluation of Cultural Competence of Vision-Language Models28 May

Other references · 11

  1. 2024A Roadmap to Pluralistic Alignmentcited by 184
  2. 2024A theory of appropriateness with applications to generative artificial intelligencecited by 25
  3. 2025Provocations from the Humanities for Generative AI Researchcited by 10
  4. 2025Meaning Is Not A Metric: Using LLMs to make cultural context legible at scalecited by 10
  5. 2025Task-Dependent Evaluation of LLM Output Homogenization: A Taxonomy-Guided Frameworkcited by 9
  6. 2024VAGUE: Visual Contexts Clarify Ambiguous Expressionscited by 5
  7. 2020Creative Captioning: An AI Grand Challenge Based on the Dixit Board Gamecited by 4
  8. 2026Ambiguity Collapse by LLMs: A Taxonomy of Epistemic Riskscited by 3
  9. 2026AI as Entertainmentcited by 2
  10. 2026AI as a Tool for Simulation-Based Experiments in Literary Studiescited by 2
  11. 2026AI Fiction in the Wildcited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Toward Socially Aware Vision-Language Models: Evaluating Cultural Competence Through Multimodal Story Generation
  2. 2025Towards Understanding Ambiguity Resolution in Multimodal Inference of Meaning
  3. 2024Can visual language models resolve textual ambiguity with visual cues? Let visual puns tell you!
  4. 2024Cracking the Code of Juxtaposition: Can AI Models Understand the Humorous Contradictions
  5. 2024RAcQUEt: Unveiling the Dangers of Overlooked Referential Ambiguity in Visual LLMs
  6. 2026Beyond Instrumental and Substitutive Paradigms: Introducing Machine Culture as an Emergent Phenomenon in Large Language Models
  7. 2026CM2: Multimodal Cultural Reasoning via an Integrated Multi-Agent Framework
  8. 2026From annotation to reasoning: Culture in language models