Copertina

Mitigating Visual Context Degradation in Large Multimodal Models: A Training-Free Decoupled Agentic Framework

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Jia, Hongrui, Jiang, Chaoya, Zhang, Shikun, Ye, Wei
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Computer Vision and Pattern Recognition
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

From Blind Spots to Gains: Diagnostic-Driven Iterative Training for Large Multimodal Models
di: Jia, Hongrui, et al.
Pubblicazione: (2026)

MaVEn: An Effective Multi-granularity Hybrid Visual Encoding Framework for Multimodal Large Language Model
di: Jiang, Chaoya, et al.
Pubblicazione: (2024)

SymDPO: Boosting In-Context Learning of Large Multimodal Models with Symbol Demonstration Direct Preference Optimization
di: Jia, Hongrui, et al.
Pubblicazione: (2024)

EVE: Verifiable Self-Evolution of MLLMs via Executable Visual Transformations
di: Heng, Yongrui, et al.
Pubblicazione: (2026)

Hallucination Augmented Contrastive Learning for Multimodal Large Language Model
di: Jiang, Chaoya, et al.
Pubblicazione: (2023)

VLM-R$^3$: Region Recognition, Reasoning, and Refinement for Enhanced Multimodal Chain-of-Thought
di: Jiang, Chaoya, et al.
Pubblicazione: (2025)

TiMix: Text-aware Image Mixing for Effective Vision-Language Pre-training
di: Jiang, Chaoya, et al.
Pubblicazione: (2023)

IKOD: Mitigating Visual Attention Degradation in Large Vision-Language Models
di: Yang, Jiabing, et al.
Pubblicazione: (2025)

OSWorld-MCP: Benchmarking MCP Tool Invocation In Computer-Use Agents
di: Jia, Hongrui, et al.
Pubblicazione: (2025)

ControlMLLM: Training-Free Visual Prompt Learning for Multimodal Large Language Models
di: Wu, Mingrui, et al.
Pubblicazione: (2024)

BUS:Efficient and Effective Vision-language Pre-training with Bottom-Up Patch Summarization
di: Jiang, Chaoya, et al.
Pubblicazione: (2023)

Training-Free Mitigation of Language Reasoning Degradation After Multimodal Instruction Tuning
di: Ratzlaff, Neale, et al.
Pubblicazione: (2024)

TRIPS: Efficient Vision-and-Language Pre-training with Text-Relevant Image Patch Selection
di: Jiang, Chaoya, et al.
Pubblicazione: (2023)

Efficient Vision-and-Language Pre-training with Text-Relevant Image Patch Selection
di: Ye, Wei, et al.
Pubblicazione: (2024)

MIBench: Evaluating Multimodal Large Language Models over Multiple Images
di: Liu, Haowei, et al.
Pubblicazione: (2024)

Echo-α: Large Agentic Multimodal Reasoning Model for Ultrasound Interpretation
di: Zhang, Jing, et al.
Pubblicazione: (2026)

Personal Visual Context Learning in Large Multimodal Models
di: Xue, Zihui, et al.
Pubblicazione: (2026)

GAMBIT: A Gamified Jailbreak Framework for Multimodal Large Language Models
di: Hu, Xiangdong, et al.
Pubblicazione: (2026)

See It, Say It, Sorted: An Iterative Training-Free Framework for Visually-Grounded Multimodal Reasoning in LVLMs
di: Zhang, Yongchang, et al.
Pubblicazione: (2026)

Connecting the Dots: Training-Free Visual Grounding via Agentic Reasoning
di: Luo, Liqin, et al.
Pubblicazione: (2025)

Visual Attention Drifts,but Anchors Hold:Mitigating Hallucination in Multimodal Large Language Models via Cross-Layer Visual Anchors
di: Yang, Chengxu, et al.
Pubblicazione: (2026)

Decoupling Training-Free Guided Diffusion by ADMM
di: Zhang, Youyuan, et al.
Pubblicazione: (2024)

Look Twice Before You Answer: Memory-Space Visual Retracing for Hallucination Mitigation in Multimodal Large Language Models
di: Zou, Xin, et al.
Pubblicazione: (2024)

DeepScan: A Training-Free Framework for Visually Grounded Reasoning in Large Vision-Language Models
di: Li, Yangfu, et al.
Pubblicazione: (2026)

Look Carefully: Adaptive Visual Reinforcements in Multimodal Large Language Models for Hallucination Mitigation
di: Zhu, Xingyu, et al.
Pubblicazione: (2026)

CRoPS: A Training-Free Hallucination Mitigation Framework for Vision-Language Models
di: Anand, Neeraj, et al.
Pubblicazione: (2026)

Extracting Visual Facts from Intermediate Layers for Mitigating Hallucinations in Multimodal Large Language Models
di: Zhou, Haoran, et al.
Pubblicazione: (2025)

Training-Free Image Editing with Visual Context Integration and Concept Alignment
di: Song, Rui, et al.
Pubblicazione: (2026)

Unleashing the Intrinsic Visual Representation Capability of Multimodal Large Language Models
di: Li, Hengzhuang, et al.
Pubblicazione: (2025)

DGL-RSIS: Decoupling Global Spatial Context and Local Class Semantics for Training-Free Remote Sensing Image Segmentation
di: Li, Boyi, et al.
Pubblicazione: (2025)

Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models
di: Wang, Wenbin, et al.
Pubblicazione: (2024)

Parameter-efficient Tuning of Large-scale Multimodal Foundation Model
di: Wang, Haixin, et al.
Pubblicazione: (2023)

DDFusion:Degradation-Decoupled Fusion Framework for Robust Infrared and Visible Images Fusion
di: Zhang, Tianpei, et al.
Pubblicazione: (2025)

ModuSeg: Decoupling Object Discovery and Semantic Retrieval for Training-Free Weakly Supervised Segmentation
di: He, Qingze, et al.
Pubblicazione: (2026)

Bridging the Gap Between Sparsity and Redundancy: A Dual-Decoding Framework with Global Context for Map Inference
di: Shen, Yudong, et al.
Pubblicazione: (2025)

FUSE: Label-Free Image-Event Joint Monocular Depth Estimation via Frequency-Decoupled Alignment and Degradation-Robust Fusion
di: Sun, Pihai, et al.
Pubblicazione: (2025)

Breaking Degradation Coupling: A Structural Entropy Guided Decoupled Framework and Benchmark for Infrared Enhancement
di: Li, Pu, et al.
Pubblicazione: (2026)

QVAD: A Question-Centric Agentic Framework for Efficient and Training-Free Video Anomaly Detection
di: Bekit, Lokman, et al.
Pubblicazione: (2026)

Mitigating Hallucinations in Large Vision-Language Models without Performance Degradation
di: Zhu, Xingyu, et al.
Pubblicazione: (2026)

TrimTokenator-LC: Towards Adaptive Visual Token Pruning for Large Multimodal Models with Long Contexts
di: Zhang, Hao, et al.
Pubblicazione: (2025)