Image de couverture de livre

Two Causes, Not One: Rethinking Omission and Fabrication Hallucinations in MLLMs

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Si, Guangzong, Yin, Hao, Li, Xianfei, Ding, Qing, Liao, Wenlong, He, Tao, Peng, Pai
Format:	Preprint
Publié:	2025
Sujets:	Computer Vision and Pattern Recognition
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

The Mirage of Performance Gains: Why Contrastive Decoding Fails to Mitigate Object Hallucinations in MLLMs?
par: Yin, Hao, et autres
Publié: (2025)

Lifting the Veil on Visual Information Flow in MLLMs: Unlocking Pathways to Faster Inference
par: Yin, Hao, et autres
Publié: (2025)

Generative Planning with 3D-vision Language Pre-training for End-to-End Autonomous Driving
par: Li, Tengpeng, et autres
Publié: (2025)

Open-Vocabulary Object Detection via Neighboring Region Attention Alignment
par: Qiang, Sunyuan, et autres
Publié: (2024)

ClearSight: Visual Signal Enhancement for Object Hallucination Mitigation in Multimodal Large language Models
par: Yin, Hao, et autres
Publié: (2025)

Rethinking Temporal Fusion with a Unified Gradient Descent View for 3D Semantic Occupancy Prediction
par: Chen, Dubing, et autres
Publié: (2025)

You Only Click Once: Single Point Weakly Supervised 3D Instance Segmentation for Autonomous Driving
par: Jiang, Guangfeng, et autres
Publié: (2025)

The DAWN of World-Action Interactive Models
par: Lu, Hongbo, et autres
Publié: (2026)

Semantic Causality-Aware Vision-Based 3D Occupancy Prediction
par: Chen, Dubing, et autres
Publié: (2025)

ARGUS: Hallucination and Omission Evaluation in Video-LLMs
par: Rawal, Ruchit, et autres
Publié: (2025)

VisionNVS: Self-Supervised Inpainting for Novel View Synthesis under the Virtual-Shift Paradigm
par: Lu, Hongbo, et autres
Publié: (2026)

One Token, Two Fates: A Unified Framework via Vision Token Manipulation Against MLLMs Hallucination
par: Fa, Zhan, et autres
Publié: (2026)

Attention Reallocation: Towards Zero-cost and Controllable Hallucination Mitigation of MLLMs
par: Tu, Chongjun, et autres
Publié: (2025)

FREAK: A Fine-grained Hallucination Evaluation Benchmark for Advanced MLLMs
par: Yin, Zhihan, et autres
Publié: (2026)

TARS: MinMax Token-Adaptive Preference Strategy for Hallucination Reduction in MLLMs
par: Zhang, Kejia, et autres
Publié: (2025)

RedundancyLens: Revealing and Exploiting Visual Token Processing Redundancy for Efficient Decoder-Only MLLMs
par: Li, Hongliang, et autres
Publié: (2025)

Interpreting and Mitigating Hallucination in MLLMs through Multi-agent Debate
par: Lin, Zheng, et autres
Publié: (2024)

Seeing Far and Clearly: Mitigating Hallucinations in MLLMs with Attention Causal Decoding
par: Tang, Feilong, et autres
Publié: (2025)

A Real-Time On-Device Defect Detection Framework for Laser Power-Meter Sensors via Unsupervised Learning
par: Zheng, Dongqi, et autres
Publié: (2025)

Mitigating Object Hallucinations in MLLMs via Multi-Frequency Perturbations
par: Li, Shuo, et autres
Publié: (2025)

Explore the Hallucination on Low-level Perception for MLLMs
par: Sun, Yinan, et autres
Publié: (2024)

Taming Hallucinations: Boosting MLLMs' Video Understanding via Counterfactual Video Generation
par: Huang, Zhe, et autres
Publié: (2025)

Instruction-Oriented Preference Alignment for Enhancing Multi-Modal Comprehension Capability of MLLMs
par: Wang, Zitian, et autres
Publié: (2025)

On Discriminative vs. Generative classifiers: Rethinking MLLMs for Action Understanding
par: Pang, Zhanzhong, et autres
Publié: (2026)

One Model, Two Minds: Task-Conditioned Reasoning for Unified Image Quality and Aesthetic Assessment
par: Yin, Wen, et autres
Publié: (2026)

Clinical Cognition Alignment for Gastrointestinal Diagnosis with Multimodal LLMs
par: Zheng, Huan, et autres
Publié: (2026)

Mitigating Visual Hallucinations via Semantic Curriculum Preference Optimization in MLLMs
par: Li, Yuanshuai, et autres
Publié: (2025)

Diagnosing and Correcting Concept Omission in Multimodal Diffusion Transformers
par: Baek, Kanghyun, et autres
Publié: (2026)

Detecting Omissions in Geographic Maps through Computer Vision
par: Nguyen, Phuc D. A., et autres
Publié: (2024)

Ground What You See: Hallucination-Resistant MLLMs via Caption Feedback, Diversity-Aware Sampling, and Conflict Regularization
par: Pan, Miao, et autres
Publié: (2026)

Mitigating Object Hallucination in MLLMs via Data-augmented Phrase-level Alignment
par: Sarkar, Pritam, et autres
Publié: (2024)

Tell Model Where to Look: Mitigating Hallucinations in MLLMs by Vision-Guided Attention
par: Zhao, Jianfei, et autres
Publié: (2025)

ReEXplore: Improving MLLMs for Embodied Exploration with Contextualized Retrospective Experience Replay
par: Zhang, Gengyuan, et autres
Publié: (2025)

Deep Learning for Inertial Positioning: A Survey
par: Chen, Changhao, et autres
Publié: (2023)

NOAH: Benchmarking Narrative Prior driven Hallucination and Omission in Video Large Language Models
par: Lee, Kyuho, et autres
Publié: (2025)

FINER: MLLMs Hallucinate under Fine-grained Negative Queries
par: Xiao, Rui, et autres
Publié: (2026)

Improving the Reasoning of Multi-Image Grounding in MLLMs via Reinforcement Learning
par: Zhang, Bob, et autres
Publié: (2025)

Harnessing the Power of MLLMs for Transferable Text-to-Image Person ReID
par: Tan, Wentao, et autres
Publié: (2024)

When Looking Is Not Enough: Visual Attention Structure Reveals Hallucination in MLLMs
par: Cao, Fanpu, et autres
Publié: (2026)

AMBER: An LLM-free Multi-dimensional Benchmark for MLLMs Hallucination Evaluation
par: Wang, Junyang, et autres
Publié: (2023)