Cover Image

Vocabulary Hijacking in LVLMs: Unveiling Critical Attention Heads by Excluding Inert Tokens to Mitigate Hallucination

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Chen, Yangneng, Li, Junlin, Yao, Weijun, Ma, Xilai, Du, Guodong, Wang, Wenya, Li, Jing
Format:	Preprint
Published:	2026
Subjects:	Multimedia Computer Vision and Pattern Recognition
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Attention Hijackers: Detect and Disentangle Attention Hijacking in LVLMs for Hallucination Mitigation
by: Chen, Beitao, et al.
Published: (2025)

Mitigating Image Captioning Hallucinations in Vision-Language Models
by: Zhao, Fei, et al.
Published: (2025)

Segmentation-Based Attention Entropy: Detecting and Mitigating Object Hallucinations in Large Vision-Language Models
by: Song, Jiale, et al.
Published: (2026)

Mitigating Hallucination in Large Vision-Language Models via Adaptive Attention Calibration
by: Fazli, Mehrdad, et al.
Published: (2025)

STEAR: Layer-Aware Spatiotemporal Evidence Intervention for Hallucination Mitigation in Video Large Language Models
by: Fan, Linfeng, et al.
Published: (2026)

MaskCD: Mitigating LVLM Hallucinations by Image Head Masked Contrastive Decoding
by: Deng, Jingyuan, et al.
Published: (2025)

Language Bias in LVLMs: From In-Depth Analysis to Simple and Effective Mitigation
by: Chen, Yangneng, et al.
Published: (2026)

Cross-Attention Fusion of Visual and Geometric Features for Large Vocabulary Arabic Lipreading
by: Daou, Samar, et al.
Published: (2024)

Mitigating Hallucination in Multimodal Large Language Model via Hallucination-targeted Direct Preference Optimization
by: Fu, Yuhan, et al.
Published: (2024)

When Harmful Content Gets Camouflaged: Unveiling Perception Failure of LVLMs with CamHarmTI
by: Li, Yanhui, et al.
Published: (2025)

DPC: Dual-Prompt Collaboration for Tuning Vision-Language Models
by: Li, Haoyang, et al.
Published: (2025)

Hierarchical Masked Autoregressive Models with Low-Resolution Token Pivots
by: Zheng, Guangting, et al.
Published: (2025)

A Tri-Dynamic Preprocessing Framework for UGC Video Compression
by: Zhao, Fei, et al.
Published: (2025)

A Preprocessing Framework for Video Machine Vision under Compression
by: Zhao, Fei, et al.
Published: (2025)

MSCT: Differential Cross-Modal Attention for Deepfake Detection
by: Wei, Fangda, et al.
Published: (2026)

MAO: Efficient Model-Agnostic Optimization of Prompt Tuning for Vision-Language Models
by: Li, Haoyang, et al.
Published: (2025)

Decompose and Transfer: CoT-Prompting Enhanced Alignment for Open-Vocabulary Temporal Action Detection
by: Zhu, Sa, et al.
Published: (2026)

FD2Talk: Towards Generalized Talking Head Generation with Facial Decoupled Diffusion Model
by: Yao, Ziyu, et al.
Published: (2024)

Efficient Token Compression for Vision Transformer with Spatial Information Preserved
by: Mao, Junzhu, et al.
Published: (2025)

Unveiling Encoder-Free Vision-Language Models
by: Diao, Haiwen, et al.
Published: (2024)

Frozen LVLMs for Micro-Video Recommendation: A Systematic Study of Feature Extraction and Fusion
by: Sun, Huatuan, et al.
Published: (2025)

Towards Universal Modal Tracking with Online Dense Temporal Token Learning
by: Zheng, Yaozong, et al.
Published: (2025)

Towards Open-Vocabulary Audio-Visual Event Localization
by: Zhou, Jinxing, et al.
Published: (2024)

T2VParser: Adaptive Decomposition Tokens for Partial Alignment in Text to Video Retrieval
by: Li, Yili, et al.
Published: (2025)

Generative Preprocessing for Image Compression with Pre-trained Diffusion Models
by: Guo, Mengxi, et al.
Published: (2025)

Towards Open-Vocabulary Remote Sensing Image Semantic Segmentation
by: Ye, Chengyang, et al.
Published: (2024)

Mitigating Hallucinations in Multimodal LLMs via Object-aware Preference Optimization
by: Compagnoni, Alberto, et al.
Published: (2025)

Mitigating Hallucinations in Large Vision-Language Models with Instruction Contrastive Decoding
by: Wang, Xintong, et al.
Published: (2024)

Hallu-PI: Evaluating Hallucination in Multi-modal Large Language Models within Perturbed Inputs
by: Ding, Peng, et al.
Published: (2024)

Probabilistic Temporal Masked Attention for Cross-view Online Action Detection
by: Xie, Liping, et al.
Published: (2025)

DanceEditor: Towards Iterative Editable Music-driven Dance Generation with Open-Vocabulary Descriptions
by: Zhang, Hengyuan, et al.
Published: (2025)

DBDH: A Dual-Branch Dual-Head Neural Network for Invisible Embedded Regions Localization
by: Zhao, Chengxin, et al.
Published: (2024)

DetailVerifyBench: A Benchmark for Dense Hallucination Localization in Long Image Captions
by: Wang, Xinran, et al.
Published: (2026)

Where Does Vision Meet Language? Understanding and Refining Visual Fusion in MLLMs via Contrastive Attention
by: Song, Shezheng, et al.
Published: (2026)

MLLM can see? Dynamic Correction Decoding for Hallucination Mitigation
by: Wang, Chenxi, et al.
Published: (2024)

CounterVid: Counterfactual Video Generation for Mitigating Action and Temporal Hallucinations in Video-Language Models
by: Poppi, Tobia, et al.
Published: (2026)

Frequency-Assisted Adaptive Sharpening Scheme Considering Bitrate and Quality Tradeoff
by: Pang, Yingxue, et al.
Published: (2025)

Revolutionizing Text-to-Image Retrieval as Autoregressive Token-to-Voken Generation
by: Li, Yongqi, et al.
Published: (2024)

Riemann-based Multi-scale Attention Reasoning Network for Text-3D Retrieval
by: Li, Wenrui, et al.
Published: (2024)

Mitigating Hallucination in Large Multi-Modal Models via Robust Instruction Tuning
by: Liu, Fuxiao, et al.
Published: (2023)