Cover Image

DREAM: Disentangling Risks to Enhance Safety Alignment in Multimodal Large Language Models

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Liu, Jianyu, Guo, Hangyu, Duan, Ranjie, Bu, Xingyuan, He, Yancheng, Li, Shilong, Huang, Hui, Liu, Jiaheng, Wang, Yucheng, Jing, Chenchen, Qu, Xingwei, Zhang, Xiao, Tan, Yingshui, Wu, Yanan, Gu, Jihao, Li, Yangguang, Zhu, Jianke
Format:	Preprint
Published:	2025
Subjects:	Computation and Language Computer Vision and Pattern Recognition
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

GraphReader: Building Graph-based Agent to Enhance Long-Context Abilities of Large Language Models
by: Li, Shilong, et al.
Published: (2024)

2D-DPO: Scaling Direct Preference Optimization with 2-Dimensional Supervision
by: Li, Shilong, et al.
Published: (2024)

Chinese SimpleQA: A Chinese Factuality Evaluation for Large Language Models
by: He, Yancheng, et al.
Published: (2024)

Can Large Language Models Detect Errors in Long Chain-of-Thought Reasoning?
by: He, Yancheng, et al.
Published: (2025)

Adaptive Segment-level Reward: Bridging the Gap Between Action and Reward Space in Alignment
by: Li, Yanshi, et al.
Published: (2024)

Token Preference Optimization with Self-Calibrated Visual-Anchored Rewards for Hallucination Mitigation
by: Gu, Jihao, et al.
Published: (2024)

PSA-VLM: Enhancing Vision-Language Model Safety through Progressive Concept-Bottleneck-Driven Alignment
by: Liu, Zhendong, et al.
Published: (2024)

MM-BrowseComp: A Comprehensive Benchmark for Multimodal Browsing Agents
by: Li, Shilong, et al.
Published: (2025)

"See the World, Discover Knowledge": A Chinese Factuality Evaluation for Large Vision Language Models
by: Gu, Jihao, et al.
Published: (2025)

Chinese SafetyQA: A Safety Short-form Factuality Benchmark for Large Language Models
by: Tan, Yingshui, et al.
Published: (2024)

Oyster-I: Beyond Refusal -- Constructive Safety Alignment for Responsible Language Models
by: Duan, Ranjie, et al.
Published: (2025)

MESA: Improving MoE Safety Alignment via Decentralized Expertise
by: Sun, Yitong, et al.
Published: (2026)

STAIR: Improving Safety Alignment with Introspective Reasoning
by: Zhang, Yichi, et al.
Published: (2025)

AIR: Complex Instruction Generation via Automatic Iterative Refinement
by: Liu, Wei, et al.
Published: (2025)

Safety Alignment for Vision Language Models
by: Liu, Zhendong, et al.
Published: (2024)

ConceptMath: A Bilingual Concept-wise Benchmark for Measuring Mathematical Reasoning of Large Language Models
by: Wu, Yanan, et al.
Published: (2024)

Safety Alignment via Constrained Knowledge Unlearning
by: Shi, Zesheng, et al.
Published: (2025)

MSR-Align: Policy-Grounded Multimodal Alignment for Safety-Aware Reasoning in Vision-Language Models
by: Xia, Yinan, et al.
Published: (2025)

MuSC: Improving Complex Instruction Following with Multi-granularity Self-Contrastive Training
by: Huang, Hui, et al.
Published: (2025)

MT-Bench-101: A Fine-Grained Benchmark for Evaluating Large Language Models in Multi-Turn Dialogues
by: Bai, Ge, et al.
Published: (2024)

Equilibrate RLHF: Towards Balancing Helpfulness-Safety Trade-off in Large Language Models
by: Tan, Yingshui, et al.
Published: (2025)

Enhancing Vision-Language Model with Unmasked Token Alignment
by: Liu, Jihao, et al.
Published: (2024)

UP-VLA: A Unified Understanding and Prediction Model for Embodied Agent
by: Zhang, Jianke, et al.
Published: (2025)

Hierarchical Disentanglement-Alignment Network for Robust SAR Vehicle Recognition
by: Li, Weijie, et al.
Published: (2023)

Iterative Length-Regularized Direct Preference Optimization: A Case Study on Improving 7B Language Models to GPT-4 Level
by: Liu, Jie, et al.
Published: (2024)

I-SHEEP: Self-Alignment of LLM from Scratch through an Iterative Self-Enhancement Paradigm
by: Liang, Yiming, et al.
Published: (2024)

Improving Adversarial Robust Fairness via Anti-Bias Soft Label Distillation
by: Zhao, Shiji, et al.
Published: (2023)

Emulated Disalignment: Safety Alignment for Large Language Models May Backfire!
by: Zhou, Zhanhui, et al.
Published: (2024)

Adversarial Orthogonal Disentanglement for LVLM Hallucination Mitigation
by: Cheng, Ruoxi, et al.
Published: (2026)

Disentangling AI Alignment: A Structured Taxonomy Beyond Safety and Ethics
by: Baum, Kevin
Published: (2025)

D-CPT Law: Domain-specific Continual Pre-Training Scaling Law for Large Language Models
by: Que, Haoran, et al.
Published: (2024)

MM-Instruct: Generated Visual Instructions for Large Multimodal Model Alignment
by: Liu, Jihao, et al.
Published: (2024)

Inverse Reinforcement Learning with Dynamic Reward Scaling for LLM Alignment
by: Cheng, Ruoxi, et al.
Published: (2025)

TEGEE: Task dEfinition Guided Expert Ensembling for Generalizable and Few-shot Learning
by: Qu, Xingwei, et al.
Published: (2024)

DREAM: Document Reconstruction via End-to-end Autoregressive Model
by: Li, Xin, et al.
Published: (2025)

Preference Alignment for Diffusion Model via Explicit Denoised Distribution Estimation
by: Shi, Dingyuan, et al.
Published: (2024)

UAM: A Dual-Stream Perspective on Forgetting in VLA Training
by: Zhang, Jianke, et al.
Published: (2026)

MF2Summ: Multimodal Fusion for Video Summarization with Temporal Alignment
by: wang, Shuo, et al.
Published: (2025)

COIG-P: A High-Quality and Large-Scale Chinese Preference Dataset for Alignment with Human Values
by: P Team, et al.
Published: (2025)

Improving Vision-Language-Action Model with Online Reinforcement Learning
by: Guo, Yanjiang, et al.
Published: (2025)