Vocabulary Hijacking in LVLMs: Unveiling Critical Attention Heads by Excluding Inert Tokens to Mitigate Hallucination
Large Vision-Language Models (LVLMs) have achieved remarkable progress in multimodal tasks, yet their reliability is persistently undermined by hallucinations-generating text that contradicts visual input. Recent studies often attribute these errors to inadequate visual attention. In this work, we analyze the attention mechanisms via the logit lens, uncovering a distinct anomaly we term Vocabulary Hijacking. We discover that specific visual tokens, defined as Inert Tokens, disproportionately attract attention. Crucially, when their intermediate hidden states are projected into the vocabulary space, they consistently decode to a fixed set of unrelated words (termed Hijacking Anchors) across layers, revealing a rigid semantic collapse. Leveraging this semantic rigidity, we propose Hijacking Anchor-Based Identification (HABI), a robust strategy to accurately localize these Inert Tokens. To quantify the impact of this phenomenon, we introduce the Non-Hijacked Visual Attention Ratio (NHAR), a novel metric designed to identify attention heads that remain resilient to hijacking and are critical for factual accuracy. Building on these insights, we propose Hijacking-Aware Visual Attention Enhancement (HAVAE), a training-free intervention that selectively strengthens the focus of these identified heads on salient visual content. Extensive experiments across multiple benchmarks demonstrate that HAVAE significantly mitigates hallucinations with no additional computational overhead, while preserving the model's general capabilities. Our code is publicly available at https://github.com/lab-klc/HAVAE.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Attention Hijackers: Detect and Disentangle Attention Hijacking in LVLMs for Hallucination Mitigation
Despite their success, Large Vision-Language Models (LVLMs) remain vulnerable to hallucinations. While existing studies attribute the cause of hallucinations to insufficient visual attention to image tokens, our findings…
AttributeDisentanglementHallucinationEmpirical Analysis of Large Vision-Language Models against Goal Hijacking via Visual Prompt Injection
We explore visual prompt injection (VPI) that maliciously exploits the ability of large vision-language models (LVLMs) to follow instructions drawn onto the input image. We propose a new VPI method, "goal hijacking via v…
Instruction FollowingStructure-Aware Robust Fine-Tuning: Defending Vision-Language-Action Robots Against Physical Attention Hijacking
Vision-Language-Action (VLA) policies promise general robotic manipulation, but their robustness against physical-world attacks remains fragile. In particular, we show that physically realizable adversarial patches can r…
When Harmful Content Gets Camouflaged: Unveiling Perception Failure of LVLMs with CamHarmTI
Large vision-language models (LVLMs) are increasingly used for tasks where detecting multimodal harmful content is crucial, such as online content moderation. However, real-world harmful content is often camouflaged, rel…
Scene UnderstandingVisual ReasoningAttention Hijacking in Trojan Transformers
Trojan attacks pose a severe threat to AI systems. Recent works on Transformer models received explosive popularity and the self-attentions are now indisputable. This raises a central question: Can we reveal the Trojans …