← Back to all news
arXiv AI By Ziqi Wen, Parsa Madinei, Miguel P. Eckstein

Revealing the Gap in Human and VLM Scene Description through Counterfactual Semantic Saliency

Read the original on arXiv AI →

The paper introduces Counterfactual Semantic Saliency (CSS), a black-box, model-agnostic method that measures how the removal of objects from a scene shifts the semantic description produced by vision‑language models (VLMs). Using CSS, the authors compare prominent VLMs to a large human psychophysics baseline (16,603 responses across 307 scenes) and find that VLMs over‑rely on large, centrally located, and highly salient objects, while under‑rely on people compared to humans. The study highlights a significant scene‑comprehension gap between VLMs and human perception, with size bias emerging as a key factor driving model‑human divergence.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Related stories

arXiv Computer Vision
Sep 28

OpenVAM: Open-World Visual Attention Modeling with VLMs

OpenVAM is a new framework for visual attention modeling that combines a dense saliency map with language‑based explanations. It uses a decoupled design: a visual pathway for precise localization and a vision‑language head that generates grounded what/why explanations. The method is trained in three stages to preserve localization while adding language grounding, and a scalable pipeline creates multi‑domain annotations for evaluation.

By Kiana Hooshanfar, Amirhossein Kazerouni, Alireza Hosseini, Michael Brudno, Babak Taati
roboticssafety
More like this →
arXiv AI
Jun 24

When Language Overwrites Vision: Over-Alignment and Geometric Debiasing in Vision-Language Models

arXiv:2605. 08245v4 Announce Type: replace-cross Abstract: Vision-Language Models (VLMs) increasingly power high-stakes applications, from medical imaging to autonomous systems, yet they routinely hallucinate, confidently describing content not present in the input.

By Harshvardhan Saini, Samyak Jha, Yiming Tang, Dianbo Liu
llmsragagentsfine-tuningmultimodalbenchmarkssafety
More like this →
arXiv AI
Sep 2

Visual Attention Faithfulness in Vision-Language Models is Heterogeneous

The study investigates whether attention weights in Vision‑Language Models (VLMs) accurately reflect model reasoning for visual inputs. Using causal perturbation analysis, it identifies three distinct processing modes—Faithful‑Sufficient, Faithful‑Distributed, and Non‑Focal—indicating heterogeneous visual attention faithfulness. The research also shows that human‑annotated ground‑truth regions align with model attention in only about 60% of cases, highlighting a systematic divergence between model visual reliance and human intuition across VQA, document, and chart tasks.

By Xurui Song, Weishi Wang, Zhongqi Yue, Kuluhan Binici, Tao Bai, Hongxin Shao, Daniel Dahlmeier, Jun Luo
llmsnlpmultimodal
More like this →