Challenge: Medical large language models exhibit high domain specificity and condensed semantics, making them vulnerable to diagnostic errors in real-world clinical settings.
Approach: They propose a framework for modeling and inducing multi-turn medical semantic jailbreaks in clinical dialogues.
Outcome: Experiments on chest X-ray-based multimodal medical dialogues show that MSIA outperforms existing jailbreak methods with an average success rate of 76.67%.

Similar Papers

Reasoning-Augmented Conversation for Multi-Turn Jailbreak Attacks on Large Language Models (2025.findings-emnlp)

Copied to clipboard

Challenge: Existing methods for implementing multi-turn jailbreaks struggle to balance semantic coherence with attack effectiveness, resulting in benign semantic drift or ineffective detection evasion.
Approach: They propose a framework that reformulates harmful queries into benign reasoning tasks and leverages LLMs’ strong reasoning capabilities to compromise safety alignment.
Outcome: The proposed framework achieves state-of-the-art attack effectiveness in complex conversational scenarios, with average ASRs increasing by up to 96%.
From LLMs to MLLMs: Exploring the Landscape of Multimodal Jailbreaking (2024.emnlp-main)

Copied to clipboard

Challenge: Recent advances in Large Language Models (LLMs) have demonstrated remarkable performance across various tasks, effectively following instructions to meet diverse user needs.
Approach: They propose a framework for evaluation benchmarks and attack techniques for LLMs and MLLMs to enhance their security.
Outcome: The proposed frameworks have been exploited to exploit the weaknesses of LLMs and MLLMs.
Reasoning Is Not All You Need: Examining LLMs for Multi-Turn Mental Health Conversations (2026.acl-long)

Copied to clipboard

Challenge: Existing evaluation frameworks focus on diagnostic accuracy and win-rates and often overlook alignment with patient-specific goals, values, and personalities required for meaningful conversations.
Approach: They propose a framework for synthetically generating realistic, multi-turn mental health sensemaking conversations and a dataset to examine their models in healthcare settings.
Outcome: The proposed framework synthesizes a dataset comprising over 2,200 patient–LLM conversations and evaluates them using human-centric criteria.
DrAgent: Empowering Large Language Models as Medical Agents for Multi-hop Medical Reasoning (2025.findings-emnlp)

Copied to clipboard

Challenge: commercial LLMs can be difficult to use in real-world clinical decision-making . a lightweight LLM can be used to collaborate with diverse clinical tools .
Approach: They propose a lightweight LLM that can be used to build medical LLMs as agents . they use recursive curriculum learning to optimize the LLM in an easy-to-hard progression .
Outcome: The proposed approach outperforms human experts in medical examinations on diverse datasets.
GAMBIT: A Gamified Jailbreak Framework for Multimodal Large Language Models (2026.acl-long)

Copied to clipboard

Challenge: Existing attacks focus on increasing the complexity of the modified visual task and do not explicitly leverage the model’s own reasoning incentives.
Approach: They propose a framework that decomposes and reassembles harmful visual semantics and constructs a gamified scene that drives the model to explore, reconstruct intent and answer as part of winning the game.
Outcome: Experiments on reasoning and non-reasoning MLLMs show that the proposed framework outperforms baseline models on both vision and text.
SemanticCamo: Jailbreaking Large Language Models through Semantic Camouflage (2025.findings-acl)

Copied to clipboard

Challenge: Large Language Models (LLMs) have made safety issues of LLMs more prominent and critical.
Approach: They propose a framework which attacks LLMs through semantic camouflage and replaces unsafe content with semantic features to conceal malicious intent .
Outcome: The proposed framework outperforms existing models in over 80% of cases and is highly effective against various defenses.
Pattern Enhanced Multi-Turn Jailbreaking: Exploiting Structural Vulnerabilities in Large Language Models (2026.findings-acl)

Copied to clipboard

Challenge: Existing multi-turn methods for large language models exploit conversational context to bypass safety constraints gradually.
Approach: They propose a framework of five conversation patterns to construct multi-turn jailbreaks through natural dialogue.
Outcome: The proposed framework exploits conversational contexts to construct multi-turn jailbreaks . it reveals that models exhibit distinct weakness profiles and model families share similar failure modes .
MultiCodeAttack: Iterative Jailbreak Attacking on LLMs with Multi-Code Prompt Injection (2026.findings-acl)

Copied to clipboard

Challenge: Existing approaches to jailbreak rely on fixed template design and a single programming language . however, existing approaches do not consider language diversity or adaptive template evolution .
Approach: They propose a structured jailbreak framework that explores and optimizes multi-language code templates.
Outcome: The proposed framework outperforms existing jailbreak baselines and produces higher harmful outputs than baseline methods.
When Background Matters: Breaking Medical Vision Language Models by Transferable Attack (2026.acl-long)

Copied to clipboard

Challenge: Existing medical attacks focus on secondary objectives such as model stealing or adversarial fine-tuning, while transferable attacks from natural images introduce visible distortions that clinicians can easily detect. Existing transferable adversarials are less effective in the medical domain.
Approach: They propose a highly transferable black-box multimodal attack that induces incorrect yet clinically plausible diagnoses while keeping perturbations imperceptible.
Outcome: The proposed method induces incorrect yet clinically plausible diagnoses while keeping perturbations imperceptible.
MUSE: MCTS-Driven Red Teaming Framework for Enhanced Multi-Turn Dialogue Safety in Large Language Models (2025.emnlp-main)

Copied to clipboard

Challenge: Existing defenses target single-turn attacks, but real-world usage involves multi-turn dialogues, exposing models to attacks that exploit conversational context to bypass safety measures.
Approach: They propose a framework that tackles multi-turn jailbreaks from both attack and defense angles.
Outcome: Experiments on large language models show that MUSE effectively mitigates multi-turn jailbreaks.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations