Mitigating Entangled Steering in Large Vision-Language Models for Hallucination Reduction

• URL: http://arxiv.org/abs/2604.07914v1
• Date: Thu, 09 Apr 2026 07:31:27 GMT
• Title: Mitigating Entangled Steering in Large Vision-Language Models for Hallucination Reduction
• Abstract summary: Large Vision-Language Models (LVLMs) have achieved remarkable success across cross-modal tasks but remain hindered by hallucinations.<n>Existing methods mitigate hallucinations but often alter generation behavior, resulting in shorter outputs and shifted token distributions.<n>We propose MESA, an effective plug-and-play framework that performs controlled and selective latent intervention for hallucination mitigation.
• Score: 49.96701537295129
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large Vision-Language Models (LVLMs) have achieved remarkable success across cross-modal tasks but remain hindered by hallucinations, producing textual outputs inconsistent with visual content. Existing methods mitigate hallucinations but often alter generation behavior, resulting in shorter outputs and shifted token distributions, especially in latent space steering approaches. We identify that this issue stems from entangled steering signals, where suppressing hallucinations inadvertently disrupts the model's intrinsic generation behavior. To address this, we propose MESA, an effective plug-and-play framework that performs controlled and selective latent intervention for hallucination mitigation. Specifically, MESA targets hallucination-relevant responses while preserving the model's original token distribution, enabling effective hallucination reduction without compromising generation behavior. Extensive experiments across diverse generative and discriminative benchmarks demonstrate that MESA consistently reduces hallucinations while better preserving generation behavior, outperforming prior methods across multiple LVLM families.

Related papers

• Rethinking Visual Neglect: Steering via Context-Preference for MLLM Hallucination Mitigation [5.041079621345155]
  We argue that the image, as a context, simultaneously competes with the model's parametric knowledge and the textual context.<n>We propose a training-free framework, Context-Preference Steering (CAS), which extracts two semantically distinct Context Preference Vectors.<n>Experiments show that CAS substantially mitigates object hallucinations without increasing residual latency and preserves native text-generation quality.
  arXiv  Detail & Related papers  (2026-05-27T05:33:06Z)
• VCE: A zero-cost hallucination mitigation method of LVLMs via visual contrastive editing [70.82867621856968]
  Large vision-language models (LVLMs) frequently suffer from Object Hallucination (OH)<n>Recent studies suggest that the hallucination problem may stem from language priors.<n>We propose Visual Contrastive Editing (VCE), a novel post-hoc method that identifies and suppresses hallucinatory tendencies.
  arXiv  Detail & Related papers  (2026-04-21T12:40:07Z)
• DHI: Leveraging Diverse Hallucination Induction for Enhanced Contrastive Factuality Control in Large Language Models [33.2779808039684]
  Large language models (LLMs) frequently produce inaccurate or fabricated information, known as "hallucinations"<n>We propose DHI (Diverse Hallucination Induction), a novel training framework that generates a broader range of hallucinations without relying on pre-annotated data.<n> DHI achieves significant performance gains over other contrastive decoding-based approaches across multiple hallucination benchmarks.
  arXiv  Detail & Related papers  (2026-01-03T10:55:41Z)
• Review of Hallucination Understanding in Large Language and Vision Models [65.29139004945712]
  We present a framework for characterizing both image and text hallucinations across diverse applications.<n>Our investigations reveal that hallucinations often stem from predictable patterns in data distributions and inherited biases.<n>This survey provides a foundation for developing more robust and effective solutions to hallucinations in real-world generative AI systems.
  arXiv  Detail & Related papers  (2025-09-26T09:23:08Z)
• Exposing Hallucinations To Suppress Them: VLMs Representation Editing With Generative Anchors [8.089908150148554]
  Multimodal large language models (MLLMs) have achieved remarkable success across diverse vision-language tasks.<n>MLLMs are highly susceptible to hallucinations, producing content that is fluent but inconsistent with visual evidence.<n>We propose a training-free, self-supervised method for hallucination mitigation.
  arXiv  Detail & Related papers  (2025-09-26T07:24:28Z)
• Mitigating Hallucinations in Large Vision-Language Models by Self-Injecting Hallucinations [73.37711261605271]
  hallucination mitigation methods are mainly based on preference alignment and require external human annotations or auxiliary models for preference data collection.<n>We propose Autonomous Preference Alignment via Self-Injection (APASI), a novel and generalizable method that mitigates hallucinations without external dependencies.<n>APASI leverages the target LVLM to self-inject hallucinations into a generated response, creating a pair of responses with varying preference levels.
  arXiv  Detail & Related papers  (2025-09-14T14:26:53Z)
• Mitigating Hallucination in VideoLLMs via Temporal-Aware Activation Engineering [83.63437999696954]
  hallucination in large language models (MLLMs) persists as a significant and under-addressed challenge in the video domain.<n>We propose a temporal-aware activation engineering framework for VideoLLMs, which adaptively identifies and manipulates hallucination-sensitive modules.
  arXiv  Detail & Related papers  (2025-05-19T08:12:06Z)
• Alleviating Hallucinations in Large Vision-Language Models through Hallucination-Induced Optimization [123.54980913741828]
  Large Visual Language Models (LVLMs) have demonstrated exceptional abilities in understanding multimodal data.<n>They invariably suffer from hallucinations, leading to a disconnect between the generated text and the corresponding images.<n>Almost all current visual contrastive decoding methods attempt to mitigate these hallucinations by introducing visual uncertainty information.<n>However, they struggle to precisely induce the hallucinatory tokens, which severely limits their effectiveness in mitigating hallucinations.
  arXiv  Detail & Related papers  (2024-05-24T08:46:31Z)
• Mitigating Hallucinations in Large Vision-Language Models with Instruction Contrastive Decoding [25.489832294197797]
  This paper introduces the Instruction Contrastive Decoding (ICD) method, a novel approach designed to reduce hallucinations during LVLM inference. Our method is inspired by our observation that what we call disturbance instructions significantly exacerbate hallucinations in multimodal fusion modules.
  arXiv  Detail & Related papers  (2024-03-27T16:04:47Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.