Cover Image

MTSA: Multi-turn Safety Alignment for LLMs through Multi-round Red-teaming

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Guo, Weiyang, Li, Jing, Wang, Wenya, LI, YU, He, Daojing, Yu, Jun, Zhang, Min
Format:	Preprint
Published:	2025
Subjects:	Cryptography and Security Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

SWA-LDM: Toward Stealthy Watermarks for Latent Diffusion Models
by: Yang, Zhonghao, et al.
Published: (2025)

Backdoors in RLVR: Jailbreak Backdoors in LLMs From Verifiable Reward
by: Guo, Weiyang, et al.
Published: (2026)

Context-Aware Hierarchical Learning: A Two-Step Paradigm towards Safer LLMs
by: Ma, Tengyun, et al.
Published: (2025)

Multi-Stream Perturbation Attack: Breaking Safety Alignment of Thinking LLMs Through Concurrent Task Interference
by: Yang, Fan
Published: (2026)

Red-teaming the Multimodal Reasoning: Jailbreaking Vision-Language Models via Cross-modal Entanglement Attacks
by: Yan, Yu, et al.
Published: (2026)

Multi-turn Jailbreaking Attack in Multi-Modal Large Language Models
by: Das, Badhan Chandra, et al.
Published: (2026)

Targeting Alignment: Extracting Safety Classifiers of Aligned LLMs
by: Ferrand, Jean-Charles Noirot, et al.
Published: (2025)

Holistic Automated Red Teaming for Large Language Models through Top-Down Test Case Generation and Multi-turn Interaction
by: Zhang, Jinchuan, et al.
Published: (2024)

Immune: Improving Safety Against Jailbreaks in Multi-modal LLMs via Inference-Time Alignment
by: Ghosal, Soumya Suvra, et al.
Published: (2024)

AgentVigil: Generic Black-Box Red-teaming for Indirect Prompt Injection against LLM Agents
by: Wang, Zhun, et al.
Published: (2025)

Reimagining Safety Alignment with An Image
by: Xia, Yifan, et al.
Published: (2025)

A Reward-driven Automated Webshell Malicious-code Generator for Red-teaming
by: Ding, Yizhong
Published: (2025)

ART: Automatic Red-teaming for Text-to-Image Models to Protect Benign Users
by: Li, Guanlin, et al.
Published: (2024)

LeakAgent: RL-based Red-teaming Agent for LLM Privacy Leakage
by: Nie, Yuzhou, et al.
Published: (2024)

SafeThinker: Reasoning about Risk to Deepen Safety Beyond Shallow Alignment
by: Fang, Xianya, et al.
Published: (2026)

Defending MoE LLMs against Harmful Fine-Tuning via Safety Routing Alignment
by: Kim, Jaehan, et al.
Published: (2025)

Safety Geometry Collapse in Multimodal LLMs and Adaptive Drift Correction
by: Guo, Jiahe, et al.
Published: (2026)

On the Feasibility of Using MultiModal LLMs to Execute AR Social Engineering Attacks
by: Bi, Ting, et al.
Published: (2025)

State-Dependent Safety Failures in Multi-Turn Language Model Interaction
by: Li, Pengcheng, et al.
Published: (2026)

SAGE-RT: Synthetic Alignment data Generation for Safety Evaluation and Red Teaming
by: Kumar, Anurakt, et al.
Published: (2024)

SecBench: A Comprehensive Multi-Dimensional Benchmarking Dataset for LLMs in Cybersecurity
by: Jing, Pengfei, et al.
Published: (2024)

Agent Safety Alignment via Reinforcement Learning
by: Sha, Zeyang, et al.
Published: (2025)

Bidirectional Intention Inference Enhances LLMs' Defense Against Multi-Turn Jailbreak Attacks
by: Tong, Haibo, et al.
Published: (2025)

A Red Teaming Roadmap Towards System-Level Safety
by: Wang, Zifan, et al.
Published: (2025)

Learning to Watermark: A Selective Watermarking Framework for Large Language Models via Multi-Objective Optimization
by: Wang, Chenrui, et al.
Published: (2025)

Atoxia: Red-teaming Large Language Models with Target Toxic Answers
by: Du, Yuhao, et al.
Published: (2024)

Measuring Safety Alignment Effects in Autonomous Security Agents
by: David, Isaac, et al.
Published: (2026)

Refusal Falls off a Cliff: How Safety Alignment Fails in Reasoning?
by: Yin, Qingyu, et al.
Published: (2025)

Incalmo: An Autonomous LLM-assisted System for Red Teaming Multi-Host Networks
by: Singer, Brian, et al.
Published: (2025)

Co-Evolutionary Multi-Modal Alignment via Structured Adversarial Evolution
by: Shi, Guoxin, et al.
Published: (2026)

Attack the Messages, Not the Agents: A Multi-round Adaptive Stealthy Tampering Framework for LLM-MAS
by: Yan, Bingyu, et al.
Published: (2025)

CoopGuard: Stateful Cooperative Agents Safeguarding LLMs Against Evolving Multi-Round Attacks
by: Li, Siyuan, et al.
Published: (2026)

What Matters For Safety Alignment?
by: Li, Xing, et al.
Published: (2026)

VisuoAlign: Safety Alignment of LVLMs with Multimodal Tree Search
by: Li, MingSheng, et al.
Published: (2025)

FreakOut-LLM: The Effect of Emotional Stimuli on Safety Alignment
by: Kuznetsov, Daniel, et al.
Published: (2026)

CORVUS: Red-Teaming Hallucination Detectors via Internal Signal Camouflage in Large Language Models
by: Min, Nay Myat, et al.
Published: (2026)

Trojan Activation Attack: Red-Teaming Large Language Models using Activation Steering for Safety-Alignment
by: Wang, Haoran, et al.
Published: (2023)

Multi-Agent Optimization for Safety Analysis of Cyber-Physical Systems: Position Paper
by: Gürcan, Önder, et al.
Published: (2024)

QGuard:Question-based Zero-shot Guard for Multi-modal LLM Safety
by: Lee, Taegyeong, et al.
Published: (2025)

Ablating Safety: Mechanisms for Removing Alignment in Language Models for Security Applications
by: David, Isaac, et al.
Published: (2026)