mlsanityData to 14 Oct 2026

Anchored, Not Graded: Vision-Language Models Fail at Slant-from-Texture Perception

arXiv 2606.06714 ↗First posted 4 Jun 2026cs.CVcited by 0

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

Abstract (arXiv)

Human perception of surface slant from texture exhibits systematic, graded biases that emerge reliably in psychophysical experiments. Prior work showed that unsupervised CNNs reproduce several human-like biases, while supervised CNNs do not. Do Vision-Language Models (VLMs) exhibit similar competences? Across multiple VLM families and model scales, zero-shot and in-context prompting both produce distinctive failures: slant is predicted at only a small set of anchors (e.g., 0degree, ±25degree, ±45degree) with little dependence on stimulus field of view, optical slant, or surface curvature. Supervised fine-tuning partially remediates the failure, but residual anchoring persists. While success in high-level vision-language benchmarks might not require sensitivity to low-level geometric cues, we interpret anchoring as a failure at the representation-to-output language interface: not necessarily an absence of geometric encoding, but a failure to express it in a graded form.

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 38 of its 64 references that are in the map. First posted on arXiv 4 Jun 2026.

Builds on

38 of 64 references.

Closest ancestors · 8

Same direction, newest first

  1. 2025Vision Language Models are Biased29 May
  2. 2025Why Vision Language Models Struggle with Visual Arithmetic? Towards Enhanced Chart and Geometry Understanding17 Feb
  3. 2024PaliGemma: A versatile 3B VLM for transfer10 Jul
  4. 2024Vision language models are blind: Failing to translate detailed visual features into words9 Jul
  5. 2024MathVerse: Does Your Multi-modal LLM Truly See the Diagrams in Visual Math Problems?21 Mar
  6. 2023Grounding Visual Illusions in Language: Do Vision-Language Models Perceive Illusions Like Humans?31 Oct
  7. 2023MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contexts3 Oct
  8. 2023Vision-Language Models for Vision Tasks: A Survey3 Apr

Other references · 30

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,403
  4. 2025Qwen2.5-VL Technical Reportcited by 5,587
  5. 2024GPT-4o System Cardcited by 4,588
  6. 2023Qwen Technical Reportcited by 3,344
  7. 2023Mistral 7Bcited by 3,164
  8. 2023MMBench: Is Your Multi-modal Model an All-around Player?cited by 2,141
  9. 2023MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGIcited by 2,080
  10. 2025Gemma 3 Technical Reportcited by 1,824
  11. 2025InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiencycited by 1,510
  12. 2024Depth Anything: Unleashing the Power of Large-Scale Unlabeled Datacited by 1,360
18 more
  1. 2016Understanding intermediate layers using linear classifier probescited by 1,353
  2. 2014Depth Map Prediction from a Single Image using a Multi-Scale Deep Networkcited by 1,274
  3. 2024Chameleon: Mixed-Modal Early-Fusion Foundation Modelscited by 927
  4. 2024Eyes Wide Shut? Exploring the Visual Shortcomings of Multimodal LLMscited by 735
  5. 2024LLaVA-NeXT-Interleave: Tackling Multi-image, Video, and 3D in Large Multimodal Modelscited by 584
  6. 2021LiT: Zero-Shot Transfer with Locked-image text Tuningcited by 577
  7. 2024SpatialRGPT: Grounded Spatial Reasoning in Vision Language Modelscited by 311
  8. 2022Visual Spatial Reasoningcited by 293
  9. 2023G-LLaVA: Solving Geometric Problem with Multi-Modal Large Language Modelcited by 210
  10. 2024SpatialBot: Precise Spatial Understanding with Vision Language Modelscited by 152
  11. 2026Ministral 3cited by 114
  12. 2024Why are Visually-Grounded Language Models Bad at Image Classification?cited by 102
  13. 2025How Far are VLMs from Visual Spatial Intelligence? A Benchmark-Driven Perspectivecited by 43
  14. 2024Anchoring Bias in Large Language Models: An Experimental Studycited by 37
  15. 2024DepthCues: Evaluating Monocular Depth Perception in Large Vision Modelscited by 19
  16. 2025Forgotten Polygons: Multimodal Large Language Models are Shape-Blindcited by 16
  17. 2026Mechanisms of Prompt-Induced Hallucination in Vision-Language Modelscited by 5
  18. 2025StreetMath: Study of LLMs' Approximation Behaviorscited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Don't Look at the Numbers: Visual Anchoring Bias and Layer-wise Representation in VLMs
  2. 2025Investigating Spatial Attention Bias in Vision-Language Models
  3. 2026Disentangling Pictorial Cue Understanding from Language Bias in VLMs via Depth Ordering Task
  4. 2024Can We Talk Models Into Seeing the World Differently?
  5. 2025The in-context inductive biases of vision-language models differ across modalities
  6. 2026Seeing Depth but Choosing Cues in Vision-Language Models
  7. 2026The Limits of Learning from Pictures and Text: Vision-Language Models and Embodied Scene Understanding
  8. 2026When Words Speak Louder than Images: Towards Understanding Language Bias in Vision-Language Models