English
Related papers

Related papers: Not All Turns Matter: Credit Assignment for Multi-…

200 papers

Attack Success Rate (ASR) evaluates each jailbreak with a single yes/no label at the end of generation, telling us whether a failure happened but not how it unfolded. Two attacks that produce equally harmful outputs may have followed…

Artificial Intelligence · Computer Science 2026-05-29 Junyoung Park , Sunghwan Park , Seongyong Ju , Jaewoo Lee

Retrieval-Augmented Generation (RAG) delivers substantial value in knowledge-intensive applications. However, its generated responses often lack transparent reasoning paths that trace back to source evidence from retrieved documents. This…

Computation and Language · Computer Science 2026-01-30 Jingyi Ren , Yekun Xu , Xiaolong Wang , Weitao Li , Ante Wang , Weizhi Ma , Yang Liu

The remarkable capabilities of Large Language Models (LLMs) have raised significant safety concerns, particularly regarding "jailbreak" attacks that exploit adversarial prompts to bypass safety alignment mechanisms. Existing defense…

Cryptography and Security · Computer Science 2025-09-30 Haibo Tong , Dongcheng Zhao , Guobin Shen , Xiang He , Dachuan Lin , Feifei Zhao , Yi Zeng

The integration of new modalities enhances the capabilities of multimodal large language models (MLLMs) but also introduces additional vulnerabilities. In particular, simple visual jailbreaking attacks can manipulate open-source MLLMs more…

Machine Learning · Computer Science 2026-03-03 Runqi Lin , Alasdair Paren , Suqin Yuan , Muyang Li , Philip Torr , Adel Bibi , Tongliang Liu

Multi-turn prompt injection follows a known attack path -- trust-building, pivoting, escalation but text-level defenses miss covert attacks where individual turns appear benign. We show this attack path leaves an activation-level signature…

Cryptography and Security · Computer Science 2026-05-01 Prashant Kulkarni

Ensuring AI safety is crucial as large language models become increasingly integrated into real-world applications. A key challenge is jailbreak, where adversarial prompts bypass built-in safeguards to elicit harmful disallowed outputs.…

Computation and Language · Computer Science 2025-03-31 Zixuan Weng , Xiaolong Jin , Jinyuan Jia , Xiangyu Zhang

Large Language Models (LLMs) have demonstrated remarkable performance across diverse tasks. Nevertheless, they still pose notable safety risks due to potential misuse for malicious purposes. Jailbreaking, which seeks to induce models to…

Computation and Language · Computer Science 2025-09-30 Hua Tang , Lingyong Yan , Yukun Zhao , Shuaiqiang Wang , Jizhou Huang , Dawei Yin

We introduce a novel framework for consolidating multi-turn adversarial ``jailbreak'' prompts into single-turn queries, significantly reducing the manual overhead required for adversarial testing of large language models (LLMs). While…

Computation and Language · Computer Science 2025-08-06 Junwoo Ha , Hyunjun Kim , Sangyoon Yu , Haon Park , Ashkan Yousefpour , Yuna Park , Suhyun Kim

Learning to compute, the ability to model the functional behavior of a circuit graph, is a fundamental challenge for graph representation learning. Yet, the dominant paradigm is architecturally mismatched for this task. This flawed…

Artificial Intelligence · Computer Science 2026-02-10 Ziyang Zheng , Jiaying Zhu , Jingyi Zhou , Qiang Xu

Many-shot jailbreaking circumvents the safety alignment of LLMs by exploiting their ability to process long input sequences. To achieve this, the malicious target prompt is prefixed with hundreds of fabricated conversational exchanges…

Computation and Language · Computer Science 2025-06-16 Avery Ma , Yangchen Pan , Amir-massoud Farahmand

Recent advances in Reinforcement Learning (RL) have underscored its potential for incentivizing reasoning capabilities of Large Language Models (LLMs). However, existing step-level efforts suffer from costly annotations that limit domain…

Machine Learning · Computer Science 2026-05-20 Junjie Zhang , Guozheng Ma , Shunyu Liu , Zetian Hu , Yongcheng Jing , Ting-En Lin , Yongbin Li , Dacheng Tao

Jailbreaks on large language models (LLMs) have recently received increasing attention. For a comprehensive assessment of LLM safety, it is essential to consider jailbreaks with diverse attributes, such as contextual coherence and…

Machine Learning · Computer Science 2024-06-10 Xingang Guo , Fangxu Yu , Huan Zhang , Lianhui Qin , Bin Hu

Post-training alignment of large language models (LLMs) relies on large-scale human annotations guided by policy specifications that change over time. Cultural shifts, value reinterpretations, and regulatory or industrial updates make…

Computation and Language · Computer Science 2026-05-12 Aakash Sen Sharma , Debdeep Sanyal , Manodeep Ray , Vivek Srivastava , Shirish Karande , Murari Mandal

As large language models (LLMs) grow in power and influence, ensuring their safety and preventing harmful output becomes critical. Automated red teaming serves as a tool to detect security vulnerabilities in LLMs without manual labor.…

Artificial Intelligence · Computer Science 2025-06-03 Weiyang Guo , Zesheng Shi , Zhuo Li , Yequan Wang , Xuebo Liu , Wenya Wang , Fangming Liu , Min Zhang , Jing Li

With the rapid advancement of large language models (LLMs), the safety of LLMs has become a critical concern. Despite significant efforts in safety alignment, current LLMs remain vulnerable to jailbreaking attacks. However, the root causes…

Artificial Intelligence · Computer Science 2026-03-10 Yonghong Deng , Zhen Yang , Ping Jian , Xinyue Zhang , Zhongbin Guo , Chengzhi Li

Multi-turn interactions with language models (LMs) pose critical safety risks, as harmful intent can be strategically spread across exchanges. Yet, the vast majority of prior work has focused on single-turn safety, while adaptability and…

Cryptography and Security · Computer Science 2025-08-26 Salman Rahman , Liwei Jiang , James Shiffer , Genglin Liu , Sheriff Issaka , Md Rizwan Parvez , Hamid Palangi , Kai-Wei Chang , Yejin Choi , Saadia Gabriel

Open-weight language models can be rendered unsafe through several distinct interventions, but the resulting models may differ substantially in capabilities, behavioral profile, and internal failure mode. We study behavioral and mechanistic…

Cryptography and Security · Computer Science 2026-04-21 Md Rysul Kabir , Zoran Tiganj

Large language models (LLMs) have demonstrated remarkable capabilities, yet they also introduce novel security challenges. For instance, prompt jailbreaking attacks involve adversaries crafting sophisticated prompts to elicit responses from…

Artificial Intelligence · Computer Science 2025-09-30 Zhaoqi Wang , Daqing He , Zijian Zhang , Xin Li , Liehuang Zhu , Meng Li , Jiamou Liu

Large Language Models (LLMs) face a significant threat from multi-turn jailbreak attacks, where adversaries progressively steer conversations to elicit harmful outputs. However, the practical effectiveness of existing attacks is undermined…

Cryptography and Security · Computer Science 2026-01-12 Songze Li , Ruishi He , Xiaojun Jia , Jun Wang , Zhihui Fu

As large language model (LLM) agents evolve from isolated tool users into coordinated teams, reinforcement learning (RL) must optimize not only individual actions but also how work is spawned, delegated, communicated, aggregated, and…

Computation and Language · Computer Science 2026-05-05 Chenchen Zhang