Imagen de Portada

Debate with Images: Detecting Deceptive Behaviors in Multimodal Large Language Models

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Fang, Sitong, Hou, Shiyi, Wang, Kaile, Chen, Boyuan, Hong, Donghai, Zhou, Jiayi, Dai, Josef, Yang, Yaodong, Ji, Jiaming
Formato:	Preprint
Publicado:	2025
Materias:	Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Mitigating Deceptive Alignment via Self-Monitoring
por: Ji, Jiaming, et al.
Publicado: (2025)

Language Models Resist Alignment: Evidence From Data Compression
por: Ji, Jiaming, et al.
Publicado: (2024)

Sequence to Sequence Reward Modeling: Improving RLHF by Language Feedback
por: Zhou, Jiayi, et al.
Publicado: (2024)

PKU-SafeRLHF: Towards Multi-Level Safety Alignment for LLMs with Human Preference
por: Ji, Jiaming, et al.
Publicado: (2024)

Generative RLHF-V: Learning Principles from Multi-modal Human Preference
por: Zhou, Jiayi, et al.
Publicado: (2025)

Stream Aligner: Efficient Sentence-Level Alignment via Distribution Induction
por: Lou, Hantao, et al.
Publicado: (2025)

Align Anything: Training All-Modality Models to Follow Instructions with Language Feedback
por: Ji, Jiaming, et al.
Publicado: (2024)

Reward Generalization in RLHF: A Topological Perspective
por: Qiu, Tianyi, et al.
Publicado: (2024)

InterMT: Multi-Turn Interleaved Preference Alignment with Human Feedback
por: Chen, Boyuan, et al.
Publicado: (2025)

Aligner: Efficient Alignment by Learning to Correct
por: Ji, Jiaming, et al.
Publicado: (2024)

Safe RLHF-V: Safe Reinforcement Learning from Multi-modal Human Feedback
por: Ji, Jiaming, et al.
Publicado: (2025)

SAE-V: Interpreting Multimodal Models for Enhanced Alignment
por: Lou, Hantao, et al.
Publicado: (2025)

SafeMT: Multi-turn Safety for Multimodal Language Models
por: Zhu, Han, et al.
Publicado: (2025)

MiraBench: Evaluating Action-Conditioned Reliability in Robotic World Models
por: Yang, Tianzhuo, et al.
Publicado: (2026)

When Slower Isn't Truer: Inverse Scaling Law of Truthfulness in Multimodal Reasoning
por: Fang, Sitong, et al.
Publicado: (2025)

SafeVLA: Towards Safety Alignment of Vision-Language-Action Model via Constrained Learning
por: Zhang, Borong, et al.
Publicado: (2025)

ThinkPatterns-21k: A Systematic Study on the Impact of Thinking Patterns in LLMs
por: Wen, Pengcheng, et al.
Publicado: (2025)

SPADE-Bench: Evaluating Spontaneous Strategic Deception in Agents via Plan-Action Divergence
por: Bu, Yuyan, et al.
Publicado: (2026)

SafeSora: Towards Safety Alignment of Text2Video Generation via a Human Preference Dataset
por: Dai, Josef, et al.
Publicado: (2024)

AI Deception: Risks, Dynamics, and Controls
por: Chen, Boyuan, et al.
Publicado: (2025)

Evolving Deception: When Agents Evolve, Deception Wins
por: Ying, Zonghao, et al.
Publicado: (2026)

AIGI-Holmes: Towards Explainable and Generalizable AI-Generated Image Detection via Multimodal Large Language Models
por: Zhou, Ziyin, et al.
Publicado: (2025)

RedVLA: Physical Red Teaming for Vision-Language-Action Models
por: Zhang, Yuhao, et al.
Publicado: (2026)

SafeLawBench: Towards Safe Alignment of Large Language Models
por: Cao, Chuxue, et al.
Publicado: (2025)

MIHBench: Benchmarking and Mitigating Multi-Image Hallucinations in Multimodal Large Language Models
por: Li, Jiale, et al.
Publicado: (2025)

Towards Explainable Harmful Meme Detection through Multimodal Debate between Large Language Models
por: Lin, Hongzhan, et al.
Publicado: (2024)

Stable Reasoning, Unstable Responses: Mitigating LLM Deception via Stability Asymmetry
por: Zhang, Guoxi, et al.
Publicado: (2026)

VISA: Value Injection via Shielded Adaptation for Personalized LLM Alignment
por: Chen, Jiawei, et al.
Publicado: (2026)

A Game-Theoretic Negotiation Framework for Cross-Cultural Consensus in LLMs
por: Zhang, Guoxi, et al.
Publicado: (2025)

SafeDreamer: Safe Reinforcement Learning with World Models
por: Huang, Weidong, et al.
Publicado: (2023)

Dynamic Emotion and Personality Profiling for Multimodal Deception Detection
por: Zheng, Li, et al.
Publicado: (2026)

Mitigating Behavioral Hallucination in Multimodal Large Language Models for Sequential Images
por: You, Liangliang, et al.
Publicado: (2025)

$γ-$MoD: Exploring Mixture-of-Depth Adaptation for Multimodal Large Language Models
por: Luo, Yaxin, et al.
Publicado: (2024)

Grounded Chain-of-Thought for Multimodal Large Language Models
por: Wu, Qiong, et al.
Publicado: (2025)

VLA-Arena: An Open-Source Framework for Benchmarking Vision-Language-Action Models
por: Zhang, Borong, et al.
Publicado: (2025)

INF-LLaVA: Dual-perspective Perception for High-Resolution Multimodal Large Language Model
por: Ma, Yiwei, et al.
Publicado: (2024)

SafeNeuron: Neuron-Level Safety Alignment for Large Language Models
por: Wang, Zhaoxin, et al.
Publicado: (2026)

Training-Free Multimodal Large Language Model Orchestration
por: Xie, Tianyu, et al.
Publicado: (2025)

SpecDiff: Accelerating Diffusion Model Inference with Self-Speculation
por: Pan, Jiayi, et al.
Publicado: (2025)

SpecEE: Accelerating Large Language Model Inference with Speculative Early Exiting
por: Xu, Jiaming, et al.
Publicado: (2025)