Image de couverture de livre

MRFD: Multi-Region Fusion Decoding with Self-Consistency for Mitigating Hallucinations in LVLMs

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Ge, Haonan, Wang, Yiwei, Yang, Ming-Hsuan, Cai, Yujun
Format:	Preprint
Publié:	2025
Sujets:	Computer Vision and Pattern Recognition Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

ChainMPQ: Interleaved Text-Image Reasoning Chains for Mitigating Relation Hallucinations
par: Wu, Yike, et autres
Publié: (2025)

CATCH: Complementary Adaptive Token-level Contrastive Decoding to Mitigate Hallucinations in LVLMs
par: Kan, Zhehan, et autres
Publié: (2024)

Cure or Poison? Embedding Instructions Visually Alters Hallucination in Vision-Language Models
par: Wang, Zhaochen, et autres
Publié: (2025)

Mitigating Object Hallucinations in LVLMs via Attention Imbalance Rectification
par: Sun, Han, et autres
Publié: (2026)

Optimizing LVLMs with On-Policy Data for Effective Hallucination Mitigation
par: Yu, Chengzhi, et autres
Publié: (2025)

PAS: A Training-Free Stabilizer for Temporal Encoding in Video LLMs
par: Sun, Bowen, et autres
Publié: (2025)

FrameMind: Frame-Interleaved Video Reasoning via Reinforcement Learning
par: Ge, Haonan, et autres
Publié: (2025)

Identify, Isolate, and Purge: Mitigating Hallucinations in LVLMs via Self-Evolving Distillation
par: Li, Wenhao, et autres
Publié: (2025)

Mitigating Hallucinations in Large Vision-Language Models (LVLMs) via Language-Contrastive Decoding (LCD)
par: Manevich, Avshalom, et autres
Publié: (2024)

Countering the Over-Reliance Trap: Mitigating Object Hallucination for LVLMs via a Self-Validation Framework
par: Liu, Shiyu, et autres
Publié: (2026)

Semantic-Aware Adaptive Visual Memory for Streaming Video Understanding
par: Wu, Hang, et autres
Publié: (2026)

CHASD: Language Increment-Calibrated Contrastive Decoding against Hallucination in LVLMs
par: Huang, Xiaoyi, et autres
Publié: (2026)

MHSA: A Lightweight Framework for Mitigating Hallucinations via Steered Attention in LVLMs
par: Ding, Wei, et autres
Publié: (2026)

Causally-Grounded Dual-Path Attention Intervention for Object Hallucination Mitigation in LVLMs
par: Yu, Liu, et autres
Publié: (2025)

Text Speaks Louder than Vision: ASCII Art Reveals Textual Biases in Vision-Language Models
par: Wang, Zhaochen, et autres
Publié: (2025)

Why LVLMs Are More Prone to Hallucinations in Longer Responses: The Role of Context
par: Zheng, Ge, et autres
Publié: (2025)

Thinking in Uncertainty: Mitigating Hallucinations in MLRMs with Latent Entropy-Aware Decoding
par: Xu, Zhongxing, et autres
Publié: (2026)

CamReasoner: Reinforcing Camera Movement Understanding via Structured Spatial Reasoning
par: Wu, Hang, et autres
Publié: (2026)

TARAC: Mitigating Hallucination in LVLMs via Temporal Attention Real-time Accumulative Connection
par: Jiang, Lei, et autres
Publié: (2025)

Attention-space Contrastive Guidance for Efficient Hallucination Mitigation in LVLMs
par: Jo, Yujin, et autres
Publié: (2026)

Steering LVLMs via Sparse Autoencoder for Hallucination Mitigation
par: Hua, Zhenglin, et autres
Publié: (2025)

Finding Distributed Object-Centric Properties in Self-Supervised Transformers
par: Rawlekar, Samyak, et autres
Publié: (2026)

Delve into Visual Contrastive Decoding for Hallucination Mitigation of Large Vision-Language Models
par: Lee, Yi-Lun, et autres
Publié: (2024)

Finding the Correct Visual Evidence Without Forgetting: Mitigating Hallucination in LVLMs via Inter-Layer Visual Attention Discrepancy
par: Xie, Yutong, et autres
Publié: (2026)

VidLBEval: Benchmarking and Mitigating Language Bias in Video-Involved LVLMs
par: Yang, Yiming, et autres
Publié: (2025)

Mitigating Cross-Image Information Leakage in LVLMs for Multi-Image Tasks
par: Park, Yeji, et autres
Publié: (2025)

What Makes LVLMs Hallucinate Less? Unveiling the Architectural Factors Behind Hallucination Robustness
par: He, Yusheng, et autres
Publié: (2026)

Mitigating Coordinate Prediction Bias from Positional Encoding Failures
par: Tao, Xingjian, et autres
Publié: (2025)

Text-Guided Layer Fusion Mitigates Hallucination in Multimodal LLMs
par: Lin, Chenchen, et autres
Publié: (2026)

DiMo-GUI: Advancing Test-time Scaling in GUI Grounding via Modality-Aware Visual Reasoning
par: Wu, Hang, et autres
Publié: (2025)

Kestrel: Grounding Self-Refinement for LVLM Hallucination Mitigation
par: Mao, Jiawei, et autres
Publié: (2026)

Seeing Clearly by Layer Two: Enhancing Attention Heads to Alleviate Hallucination in LVLMs
par: Zhang, Xiaofeng, et autres
Publié: (2024)

SDCD: Structure-Disrupted Contrastive Decoding for Mitigating Hallucinations in Large Vision-Language Models
par: Xia, Yuxuan, et autres
Publié: (2026)

Visual Description Grounding Reduces Hallucinations and Boosts Reasoning in LVLMs
par: Ghosh, Sreyan, et autres
Publié: (2024)

Decoding by Perturbation: Mitigating MLLM Hallucinations via Dynamic Textual Perturbation
par: Jia, Sihang, et autres
Publié: (2026)

Intervene-All-Paths: Unified Mitigation of LVLM Hallucinations across Alignment Formats
par: Qian, Jiaye, et autres
Publié: (2025)

Mitigating Hallucinations in Video Large Language Models via Spatiotemporal-Semantic Contrastive Decoding
par: Gao, Yuansheng, et autres
Publié: (2026)

Self-Introspective Decoding: Alleviating Hallucinations for Large Vision-Language Models
par: Huo, Fushuo, et autres
Publié: (2024)

GLSim: Detecting Object Hallucinations in LVLMs via Global-Local Similarity
par: Park, Seongheon, et autres
Publié: (2025)

Mitigating Hallucinations via Inter-Layer Consistency Aggregation in Large Vision-Language Models
par: Tang, Kai, et autres
Publié: (2025)