Paper Library

Towards Reliable and Practical LLM Security Evaluations via Bayesian Modelling

Mary Llewellyn, Annie Gray, Josh Collyer, Michael Harries

2025-10-07

red teaming

2510.05709v1

Evaluating LLM Safety Across Child Development Stages: A Simulated Agent Approach

Abhejay Murali, Saleh Afroogh, Kevin Chen, David Atkinson, Amit Dhurandhar, Junfeng Jiao

2025-10-07

safety

2510.05484v1

Adversarial Reinforcement Learning for Large Language Model Agent Safety

Zizhao Wang, Dingcheng Li, Vaishakh Keshava, Phillip Wallis, Ananth Balashankar, Peter Stone, Lukas Rutishauser

2025-10-06

red teaming

2510.05442v1

AutoDAN-Reasoning: Enhancing Strategies Exploration based Jailbreak Attacks with Test-Time Scaling

Xiaogeng Liu, Chaowei Xiao

2025-10-06

red teaming

2510.05379v2

AutoDAN-Reasoning: Enhancing Strategies Exploration based Jailbreak Attacks with Test-Time Scaling

Xiaogeng Liu, Chaowei Xiao

2025-10-06

red teaming

2510.05379v1

WaveSP-Net: Learnable Wavelet-Domain Sparse Prompt Tuning for Speech Deepfake Detection

Xi Xuan, Xuechen Liu, Wenxin Zhang, Yi-Cheng Lin, Xiaojian Lin, Tomi Kinnunen

2025-10-06

2510.05305v1

Indirect Prompt Injections: Are Firewalls All You Need, or Stronger Benchmarks?

Rishika Bhagwatkar, Kevin Kasa, Abhay Puri, Gabriel Huang, Irina Rish, Graham W. Taylor, Krishnamurthy Dj Dvijotham, Alexandre Lacoste

2025-10-06

red teaming

2510.05244v1

Proactive defense against LLM Jailbreak

Weiliang Zhao, Jinjun Peng, Daniel Ben-Levi, Zhou Yu, Junfeng Yang

2025-10-06

red teaming safety

2510.05052v1

Imperceptible Jailbreaking against Large Language Models

Kuofeng Gao, Yiming Li, Chao Du, Xin Wang, Xingjun Ma, Shu-Tao Xia, Tianyu Pang

2025-10-06

red teaming

2510.05025v1

RL Is a Hammer and LLMs Are Nails: A Simple Reinforcement Learning Recipe for Strong Prompt Injection

Yuxin Wen, Arman Zharmagambetov, Ivan Evtimov, Narine Kokhlikyan, Tom Goldstein, Kamalika Chaudhuri, Chuan Guo

2025-10-06

red teaming

2510.04885v1

Unified Threat Detection and Mitigation Framework (UTDMF): Combating Prompt Injection, Deception, and Bias in Enterprise-Scale Transformers

Santhosh KumarRavindran

2025-10-06

red teaming

2510.04528v1

P2P: A Poison-to-Poison Remedy for Reliable Backdoor Defense in LLMs

Shuai Zhao, Xinyi Wu, Shiqian Zhao, Xiaobao Wu, Zhongliang Guo, Yanhao Jia, Anh Tuan Luu

2025-10-06

safety

2510.04503v2

P2P: A Poison-to-Poison Remedy for Reliable Backdoor Defense in LLMs

Shuai Zhao, Xinyi Wu, Shiqian Zhao, Xiaobao Wu, Zhongliang Guo, Yanhao Jia, Anh Tuan Luu

2025-10-06

safety

2510.04503v1

October 06 - October 12, 2025

Towards Reliable and Practical LLM Security Evaluations via Bayesian Modelling

Evaluating LLM Safety Across Child Development Stages: A Simulated Agent Approach

Adversarial Reinforcement Learning for Large Language Model Agent Safety

AutoDAN-Reasoning: Enhancing Strategies Exploration based Jailbreak Attacks with Test-Time Scaling

AutoDAN-Reasoning: Enhancing Strategies Exploration based Jailbreak Attacks with Test-Time Scaling

WaveSP-Net: Learnable Wavelet-Domain Sparse Prompt Tuning for Speech Deepfake Detection

Indirect Prompt Injections: Are Firewalls All You Need, or Stronger Benchmarks?

Proactive defense against LLM Jailbreak

Imperceptible Jailbreaking against Large Language Models

RL Is a Hammer and LLMs Are Nails: A Simple Reinforcement Learning Recipe for Strong Prompt Injection

Unified Threat Detection and Mitigation Framework (UTDMF): Combating Prompt Injection, Deception, and Bias in Enterprise-Scale Transformers

P2P: A Poison-to-Poison Remedy for Reliable Backdoor Defense in LLMs

P2P: A Poison-to-Poison Remedy for Reliable Backdoor Defense in LLMs

September 29 - October 05, 2025

Read the Scene, Not the Script: Outcome-Aware Safety for LLMs

AgentTypo: Adaptive Typographic Prompt Injection Attacks against Black-box Multimodal Agents

COSMO-RL: Towards Trustworthy LMRMs via Joint Safety and Stability

Backdoor-Powered Prompt Injection Attacks Nullify Defense Methods

Certifiable Safe RLHF: Fixed-Penalty Constraint Optimization for Safer Language Models

NEXUS: Network Exploration for eXploiting Unsafe Sequences in Multi-Turn LLM Jailbreaks

Untargeted Jailbreak Attack

External Data Extraction Attacks against Retrieval-Augmented Large Language Models

Attack via Overfitting: 10-shot Benign Fine-tuning to Jailbreak LLMs

VeriGuard: Enhancing LLM Agent Safety via Verified Code Generation

RedCodeAgent: Automatic Red-teaming Agent against Diverse Code Agents