CORTEXA
← Browse
arxivcs.AI2026-06-28

FADE: Mitigating Hallucinations by Reducing Language-Prior Dominance in Large Vision-Language Models

Yichen Guo, Kai Tang, Fenglai Lin, Yiding Sun, Dongxu Zhang, Wenya Wang, Lin William Cong, Shanghang Zhang

Despite the impressive capabilities of Large Vision-Language Models (LVLMs), they remain susceptible to hallucination, generating content inconsistent with the input image. Recent studies attribute this to the dominance of language priors over visual inputs and employ contrastive decoding methods to mitigate this dominance, but the mechanistic origin remains unexplored. We investigate the information flow through each transformer layer and find that attention modules consistently aggregate visual evidence, while FFN modules at critical layers act as the source of language priors. These priors can override visual evidence, causing correct predictions in intermediate layers to drift toward incorrect outputs. Based on this insight, we propose FADE (FFN Attenuation for DEcoding), a training-free method that attenuates FFN outputs to reduce language-prior dominance. Evaluations on POPE, CHAIR, and MME benchmarks across LLaVA-1.5, mPLUG-Owl2, and InstructBLIP show that FADE effectively mitigates hallucinations while preserving inference efficiency.

View free PDFSource page

Related papers

arxivcs.AIcs.CLcs.HC2026-07-16

Benchmarking Multimodal Large Language Models for Scientific Visualization Literacy

Patrick Phuoc Do, Chau M. Ta, Chaoli Wang

Multimodal large language models (MLLMs) are increasingly used to interpret visualizations, yet current evaluations remain largely chart-centric and provide limited evidence of understanding of scientific visualization (SciVis). We benchmark six MLLMs on the scientific visualizat…

View free PDFSource page
arxivcs.ROcs.AI2026-07-21

Agentic Real2Sim: Physics-based World Modeling with Vision-Language Agents

Guanxiong Chen, Qianjun Xia, Jiawei Peng, Heng Zhang, Bole Ma, Justin Qian, et al.

Real-to-sim conversion for robotic interaction with objects remains labor-intensive because it requires more than visual reconstruction: a streamlined real2sim process must recover scene geometries and object states, infer physical parameters, and assemble actors, objects, camera…

View free PDFSource page
arxivcs.CLcs.AI2026-07-21

Prompt Design at Scale: How Format, Instruction Count, and Context Length Shape Instruction Adherence and Hallucination in Large Language Models

Netanel Eliav

Practitioners make three prompt-design decisions with almost no controlled evidence behind them: how to format instructions and context (markdown, plain text, prose, or tabular), how many simultaneous instructions a system prompt can carry before compliance degrades, and how much…

View free PDFSource page
arxivcs.CVcs.AIcs.LGeess.IV2026-07-17

Ask Twice, Look Twice: Prompt Echoing Resolves the Question-First Paradox in Vision-Language Models

Rakshanda Hassan Abhinandan, John Galeotti, Deva Ramanan, Gautam Rajendrakumar Gare

Where should the question go in a vision-language model (VLM) prompt: before the image or after it? Intuition says before: knowing what is asked should tell the model where to look. Yet across visual question answering benchmarks, question-first prompting consistently underperfor…

View free PDFSource page
arxivcs.CVcs.AI2026-07-23

When Are Reasoning-Based Guardrails Not Efficient? ResponseGuard: A Fast Vision-Language Guard for Real-Time Moderation

Dongbin Na

A vision-language AI assistant returns its answer as a stream of generated tokens. Therefore, a safety guard that watches that answer has to keep up with the stream and stop a harmful reply before a user reads it. Recent vision-language guardrails instead generate a chain of thou…

View free PDFSource page