Cover Image

Defensive M2S: Training Guardrail Models on Compressed Multi-turn Conversations

Fuente: arXiv

Saved in:

Bibliographic Details
Main Author:	Kim, Hyunjun
Format:	Preprint
Published:	2026
Subjects:	Computation and Language Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

X-Teaming Evolutionary M2S: Automated Discovery of Multi-turn to Single-turn Jailbreak Templates
by: Kim, Hyunjun, et al.
Published: (2025)

M2S: Multi-turn to Single-turn jailbreak in Red Teaming for LLMs
by: Ha, Junwoo, et al.
Published: (2025)

On the Multi-turn Instruction Following for Conversational Web Agents
by: Deng, Yang, et al.
Published: (2024)

Learning to Clarify: Multi-turn Conversations with Action-Based Contrastive Self-Training
by: Chen, Maximillian, et al.
Published: (2024)

MacroBench: A Novel Testbed for Web Automation Scripts via Large Language Models
by: Kim, Hyunjun, et al.
Published: (2025)

Token Statistics Reveal Conversational Drift in Multi-turn LLM Interaction
by: Hafez, Wael, et al.
Published: (2026)

Enhancing Building Semantics Preservation in AI Model Training with Large Language Model Encodings
by: Jang, Suhyung, et al.
Published: (2026)

MAC: A Multi-Agent Framework for Interactive User Clarification in Multi-turn Conversations
by: Acikgoz, Emre Can, et al.
Published: (2025)

Can a Single Model Master Both Multi-turn Conversations and Tool Use? CoALM: A Unified Conversational Agentic Language Model
by: Acikgoz, Emre Can, et al.
Published: (2025)

Broaden your SCOPE! Efficient Multi-turn Conversation Planning for LLMs with Semantic Space
by: Chen, Zhiliang, et al.
Published: (2025)

NeuroFilter: Privacy Guardrails for Conversational LLM Agents
by: Das, Saswat, et al.
Published: (2026)

Building Guardrails for Large Language Models
by: Dong, Yi, et al.
Published: (2024)

Krul: Efficient State Restoration for Multi-turn Conversations with Dynamic Cross-layer KV Sharing
by: Wen, Junyi, et al.
Published: (2025)

A Decade-Scale Benchmark Evaluating LLMs' Clinical Practice Guidelines Detection and Adherence in Multi-turn Conversations
by: Tan, Andong, et al.
Published: (2026)

RAG Makes Guardrails Unsafe? Investigating Robustness of Guardrails under RAG-style Contexts
by: She, Yining, et al.
Published: (2025)

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

Trust-Oriented Adaptive Guardrails for Large Language Models
by: Hu, Jinwei, et al.
Published: (2024)

Adversarial Attacks and Defense for Conversation Entailment Task
by: Yang, Zhenning, et al.
Published: (2024)

Deep Research with Open-Domain Evaluation and Multi-Stage Guardrails for Safety
by: Huang, Wei-Chieh, et al.
Published: (2025)

Securing Multi-turn Conversational Language Models From Distributed Backdoor Triggers
by: Tong, Terry, et al.
Published: (2024)

Stop Listening to Me! How Multi-turn Conversations Can Degrade LLM Reliability
by: Guo, Kevin H., et al.
Published: (2026)

THRD: A Training-Free Multi-Turn Defense Framework for Jailbreak Attacks on Large Language Models
by: Ma, Zhiqing, et al.
Published: (2026)

Unified Multi-Task Learning & Model Fusion for Efficient Language Model Guardrailing
by: Neill, James O', et al.
Published: (2025)

Safety Through Reasoning: An Empirical Study of Reasoning Guardrail Models
by: Sreedhar, Makesh Narsimhan, et al.
Published: (2025)

RedacBench: Can AI Erase Your Secrets?
by: Jeon, Hyunjun, et al.
Published: (2026)

TurnWise: The Gap between Single- and Multi-turn Language Model Capabilities
by: Graf, Victoria, et al.
Published: (2026)

SOMA: Efficient Multi-turn LLM Serving via Small Language Model
by: Cheng, Xueqi, et al.
Published: (2026)

MindEval: Benchmarking Language Models on Multi-turn Mental Health Support
by: Pombal, José, et al.
Published: (2025)

Mitigating Lost in Multi-turn Conversation via Curriculum RL with Verifiable Accuracy and Abstention Rewards
by: Li, Ming, et al.
Published: (2025)

BARRED: Synthetic Training of Custom Policy Guardrails via Asymmetric Debate
by: Mazza, Arnon, et al.
Published: (2026)

Data Selection for Multi-turn Dialogue Instruction Tuning
by: Li, Bo, et al.
Published: (2026)

Asymmetric Actor-Critic for Multi-turn LLM Agents
by: Jiang, Shuli, et al.
Published: (2026)

Foot-In-The-Door: A Multi-turn Jailbreak for LLMs
by: Weng, Zixuan, et al.
Published: (2025)

A Lightweight Explainable Guardrail for Prompt Safety
by: Islam, Md Asiful, et al.
Published: (2026)

Speak Out of Turn: Safety Vulnerability of Large Language Models in Multi-turn Dialogue
by: Zhou, Zhenhong, et al.
Published: (2024)

From Self-Evolving Synthetic Data to Verifiable-Reward RL: Post-Training Multi-turn Interactive Tool-Using Agents
by: Gao, Jiaxuan, et al.
Published: (2026)

Lightweight Safety Guardrails via Synthetic Data and RL-guided Adversarial Training
by: Ilin, Aleksei, et al.
Published: (2025)

ADVERSA: Measuring Multi-Turn Guardrail Degradation and Judge Reliability in Large Language Models
by: Owiredu-Ashley, Harry
Published: (2026)

Improving Multi-turn Dialogue Consistency with Self-Recall Thinking
by: Pang, Renning, et al.
Published: (2026)

Taking a turn for the better: Conversation redirection throughout the course of mental-health therapy
by: Nguyen, Vivian, et al.
Published: (2024)