Large Language Model Safety

Follow this field Ask about this field

Recent momentum

emerging

0 papers in the last 28 days · 0.0% of indexed attention

Twelve weeks of publication activity for this topic as it is defined today.

Weekly history

Recent digests

What was published in this field, kept on the site without email delivery.

Period ending 2026-09-21

27 new papers

A weekly snapshot of new work published in Large Language Model Safety.

Period ending 2026-09-14

12 new papers

A weekly snapshot of new work published in Large Language Model Safety.

Inside this field

Focused directions

Focused topic

Large Language Model Safety

Within Large Language Model Safety

532 papersExplore

Focused topic

Language Model Evasion Attacks

Within Large Language Model Safety

204 papersExplore

Focused topic

Adversarial Evaluation

Within Large Language Model Safety

114 papersExplore

Focused topic

Adversarial Prompting

Within Large Language Model Safety

30 papersExplore

855 papers

Latest in Large Language Model Safety

Browse all topics →

Steer-to-Detect: Probing Hidden Representations for Detection of LLM-Generated Texts
May 13, 2026Luxu Liang, Xiang LiMachine-Generated Text Detection Language Model Evasion Attacks
Quantifying LLM Safety Degradation Under Repeated Attacks Using Survival Analysis
May 13, 2026Zvi TopolLarge Language Model Safety Large Language Model Jailbreaks
REALISTA: Realistic Latent Adversarial Attacks that Elicit LLM Hallucinations
May 12, 2026Buyun Liang, Jinqi Luo, Liangzu Peng +6Large Language Model Hallucination Adversarial Prompting
Towards Automated Air Traffic Safety Assessment Around Non-Towered Airports Using Large Language Models
May 12, 2026Torsten Darrell, Mahyar Ghazanfari, Jordan Kam +3Large Language Model Safety Safety
Reconstruction of Personally Identifiable Information from Supervised Finetuned Models
May 12, 2026Sae Furukawa, Alina OpreaSupervised Fine-Tuning Language Model Evasion Attacks
PreScam: A Benchmark for Predicting Scam Progression from Early Conversations
May 12, 2026Weixiang Sun, Shang Ma, Yiyang Li +5Fraud Detection Language Model Evasion Attacks
Persona-Conditioned Adversarial Prompting: Multi-Identity Red-Teaming for Adversarial Discovery and Mitigation
May 12, 2026Cristian Morasso, Anisa Halimi, Muhammad Zaid Hameed +1Adversarial Prompting Red-Teaming
Toward Stable Value Alignment: Introducing Independent Modules for Consistent Value Guidance
May 12, 2026Wenhao Chen, Sirui Sun, Shengyuan Bai +1Large Language Model Alignment Large Language Model Safety
When Emotion Becomes Trigger: Emotion-style dynamic Backdoor Attack Parasitising Large Language Models
May 12, 2026Ziyu Liu, Tao Li, Tianjie Ni +5Language Model Evasion Attacks Backdoor Attacks
The Evaluation Differential: When Frontier AI Models Recognise They Are Being Tested
May 12, 2026Varad Vishwarupe, Nigel Shadbolt, Marina Jirotka +1Artificial Intelligence Evaluation Artificial Intelligence Safety
Can a Single Message Paralyze the AI Infrastructure? The Rise of AbO-DDoS Attacks through Targeted Mobius Injection
May 12, 2026Zi Liang, Ronghua Li, Yanyun Wang +2Language Model Evasion Attacks Denial-Of-Service Attacks
A Mechanistic Investigation of Supervised Fine Tuning
May 12, 2026Ruhaan ChopraSupervised Fine-Tuning Model Activations
Natural Language based Specification and Verification
May 11, 2026Zhaorui Li, Chengyu SongLarge Language Model Safety Vulnerable Code
Adversarial SQL Injection Generation with LLM-Based Architectures
May 11, 2026Ali Karakoc, H. Birkan YilmazNovel Dual-Branch Identity Injection Mechanism Language Model Evasion Attacks
Beyond Red-Teaming: Formal Guarantees of LLM Guardrail Classifiers
May 11, 2026Nikita Kezins, Urbas Ekka, Pascal Berrang +1Large Language Model Safety Toxicity
RUBEN: Rule-Based Explanations for Retrieval-Augmented LLM Systems
May 11, 2026Joel Rorseth, Parke Godfrey, Lukasz Golab +2Large Language Model Safety Minimal
From Controlled to the Wild: Evaluation of Pentesting Agents for the Real-World
May 11, 2026Pedro Conde, Henrique Branquinho, Valerio Mazzone +3Penetration Testing Vulnerability Detection & Repair
LITMUS: Benchmarking Behavioral Jailbreaks of LLM Agents in Real OS Environments
May 11, 2026Chiyu Zhang, Huiqin Yang, Bendong Jiang +8Large Language Model Safety Security Analysis
MATRA: Modeling the Attack Surface of Agentic AI Systems -- OpenClaw Case Study
May 11, 2026Tim Van hamme, Thomas Vissers, Javier Carnerero-Cano +4Threat Models Artificial Intelligence Safety
Content-Aware Attack Detection in LLM Agent Tool-Call Traffic: An Empirical Study of Features, Architectures, and Evaluation Protocols
May 11, 2026Sultan ZavrakLanguage Model Evasion Attacks Artificial Intelligence Detection
Intrinsic Guardrails: How Semantic Geometry of Personality Interacts with Emergent Misalignment in LLMs
May 11, 2026Krishak Aneja, Manas Mittal, Anmol Goel +2Harmful Fine-Tuning Large Language Model Alignment
PRISM: Generation-Time Detection and Mitigation of Secret Leakage in Multi-Agent LLM Pipelines
May 11, 2026Riya Tapwal, Abhishek Kumar, Carsten MapleData Leakage Language Model Evasion Attacks
ThreatCore: A Benchmark for Explicit and Implicit Threat Detection
May 11, 2026Davide Bruni, Carlo Bardazzi, Maurizio TesconiHate Speech Threat Models
SciIntegrity-Bench: A Benchmark for Evaluating Academic Integrity in AI Scientist Systems
May 11, 2026Zonglin Yang, Xingtong Liu, Xinyan XuAcademic Integrity Artificial Intelligence Scientists
Benchmarking Safety Risks of Knowledge-Intensive Reasoning under Malicious Knowledge Editing
May 11, 2026Qinghua Mao, Xi Lin, Jinze Gu +3Large Language Model Safety Malicious Code
FragBench: Cross-Session Attacks Hidden in Benign-Looking Fragments
May 10, 2026Astha Mehta, Niruthiha Selvanayagam, Cedric Lam +10Large Language Model Safety Transferable Attack
AgentShield: Deception-based Compromise Detection for Tool-using LLM Agents
May 10, 2026Yassin H. Rassul, Tarik A. RashidLanguage Model Evasion Attacks Deception
SwordBench: Evaluating Orthogonality of Steering Image Representations
May 10, 2026Vladimir Zaigrajew, Dawid Pludowski, Hubert Baniecki +1Vision Foundation Models Adversarial Evaluation
CHAINTRIX: A multi-pipeline LLM-augmented framework for automated smart-contract security auditing
May 10, 2026Gabriela Dobrita, Simona-Vasilica Oprea, Adela BaraSmart Contracts Large Language Model Safety
The Art of the Jailbreak: Formulating Jailbreak Attacks for LLM Security Beyond Binary Scoring
May 9, 2026Ismail Hossain, Tanzim Ahad, Md Jahangir Alam +3Jailbreak Attacks Jailbreaks
GAMBIT: A Three-Mode Benchmark for Adversarial Robustness in Multi-Agent LLM Collectives
May 9, 2026Alexandre Le Mercier, Chris Develder, Thomas DemeesterAdversarial Evaluation Artificial Intelligence Detection
LLM-Agnostic Semantic Representation Attack
May 9, 2026Jiawei Lian, Jianhong Pan, Lefan Wang +4Language Model Evasion Attacks Transformation-Based Attacks
OTora: A Unified Red Teaming Framework for Reasoning-Level Denial-of-Service in LLM Agents
May 9, 2026Xinyu Li, Ronghui Mu, Lin Li +2Language Model Evasion Attacks Large Language Model Reasoning
PASA: A Principled Embedding-Space Watermarking Approach for LLM-Generated Text under Semantic-Invariant Attacks
May 9, 2026Zhenxin Ai, Haiyun HeLarge Language Model Watermarks Watermarking
CUDABeaver: Benchmarking LLM-Based Automated CUDA Debugging
May 8, 2026Shiyang Li, Haoyang Chen, Mattia Fazzini +1Cuda Precise Debugging
GLiGuard: Schema-Conditioned Classification for LLM Safeguard
May 8, 2026Urchade Zaratiana, Mary Newhauser, George Hurn-Maloney +1Large Language Model Safety Guardrail
Graph Representation Learning Augmented Model Manipulation on Federated Fine-Tuning of LLMs
May 8, 2026Hanlin Cai, Kai Li, Houtianfu Wang +4Large Language Model Fine-Tuning Language Model Evasion Attacks
LLM Wardens: Mitigating Adversarial Persuasion with Third-Party Conversational Oversight
May 8, 2026Lennart Wachowiak, Scott D. Blain, David Williams-King +1Language Model Evasion Attacks Large Language Model Decision-Making
How Value Induction Reshapes LLM Behaviour
May 8, 2026Arnav Arora, Natalie Schluter, Katherine Metcalf +1Human Values Large Language Model Safety
Beyond "I cannot fulfill this request": Alleviating Rigid Rejection in LLMs via Label Enhancement
May 8, 2026Ying Zhang, Congyu Qiao, Xin Geng +1Large Language Model Safety Rejection
Seed Hijacking of LLM Sampling and Quantum Random Number Defense
May 8, 2026Ziyang You, Xiaoke Yang, Zhanling Fan +3Language Model Evasion Attacks Random
RuleSafe-VL: Evaluating Rule-Conditioned Decision Reasoning in Vision-Language Content Moderation
May 8, 2026Zhifeng Lu, Dianyuan Wang, Yuhu Shang +1Content Moderation Vision-Language
Vaporizer: Breaking Watermarking Schemes for Large Language Model Outputs
May 8, 2026Jonathan Hong Jin Ng, Anh Tu Ngo, Anupam ChattopadhyayLarge Language Model Watermarks Watermarking
Mitigating Many-shot Jailbreak Attacks with One Single Demonstration
May 8, 2026Kejia Chen, Jiawen Zhang, Boheng Li +6Jailbreak Attacks Jailbreak Evaluations
From Clouds to Hallucinations: Atmospheric Retrieval Hijacking in Remote Sensing Vision-Language RAG
May 8, 2026Jiaju Han, Chao Li, Chengyin Hu +8Remote Sensing Multi-Source Retrieval-Augmented Generation
MIPIAD: Multilingual Indirect Prompt Injection Attack Defense with Qwen -- TF-IDF Hybrid and Meta-Ensemble Learning
May 8, 2026Al Muhit Muhtadi, Mostafa Rifat TazwarIndirect Prompt Injection Language Model Evasion Attacks
Hard to Read, Easy to Jailbreak: How Visual Degradation Bypasses MLLM Safety Alignment
May 8, 2026Zhixue Song, Boyan Han, Yiwei Wang +1Large Language Model Safety Large Vision Language Models
Adaptive auditing of AI systems with anytime-valid guarantees
May 7, 2026Siyu Zhou, Patrick Vossler, Venkatesh Sivaraman +2Algorithm Auditing Adversarial Evaluation
When No Benchmark Exists: Validating Comparative LLM Safety Scoring Without Ground-Truth Labels
May 7, 2026Sushant Gautam, Finn Schwall, Annika Willoch Olstad +6Safety Evaluation Large Language Model Safety
FedAttr: Towards Privacy-preserving Client-Level Attribution in Federated LLM Fine-tuning
May 7, 2026Su Zhang, Junfeng Guo, Heng HuangFederated Learning Privacy-Preserving Framework
Memory Efficient Full-gradient Attacks (MEFA) Framework for Adversarial Defense Evaluations
May 7, 2026Yuan Du, Mitchel Hill, HanQin CaiAdversarial Evaluation
Correct Code, Vulnerable Dependencies: A Large Scale Measurement Study of LLM-Specified Library Versions
May 7, 2026Chengjie Wang, Jingzheng Wu, Xiang Ling +2Code Quality Large Language Model Safety
Beyond Accuracy: Policy Invariance as a Reliability Test for LLM Safety Judges
May 7, 2026Shihao Weng, Yang Feng, Xiaofei XieLlm-As-A-Judge Judges
Safety Anchor: Defending Harmful Fine-tuning via Geometric Bottlenecks
May 7, 2026Guoxin Lu, Letian Sha, Qing Wang +4Harmful Fine-Tuning Large Language Model Safety
LeakDojo: Decoding the Leakage Threats of RAG Systems
May 7, 2026Maosen Zhang, Jianshuo Dong, Boting Lu +4Retrieval-Augmented Generation Systems Language Model Evasion Attacks
SafeHarbor: Hierarchical Memory-Augmented Guardrail for LLM Agent Safety
May 7, 2026Zhe Liu, Zonghao Ying, Wenxin Zhang +5Large Language Model Safety Large Language Model Agents
PersonaTeaming: Supporting Persona-Driven Red-Teaming for Generative AI
May 7, 2026Wesley Hanwen Deng, Mingxi Yan, Sunnie S. Y. Kim +5Red-Teaming Human-Ai Collaboration
Chain of Risk: Safety Failures in Large Reasoning Models and Mitigation via Adaptive Multi-Principle Steering
May 7, 2026Xiaomin Li, Jianheng Hou, Zheyuan Deng +6Large Language Model Safety Large Reasoning Models
XL-SafetyBench: A Country-Grounded Cross-Cultural Benchmark for LLM Safety and Cultural Sensitivity
May 7, 2026Dasol Choi, Eugenia Kim, Jaewon Noh +14Large Language Model Safety Cultural Knowledge
Architecture Matters: Comparing RAG Systems under Knowledge Base Poisoning
May 7, 2026Samuel KornRetrieval-Augmented Generation Systems Adversarial Evaluation

May 13, 2026 · stat.APJ/K move · Enter open · S save

Steer-to-Detect: Probing Hidden Representations for Detection of LLM-Generated Texts

Luxu Liang, Xiang Li

Tsinghua University · University of Pennsylvania

The rapid advancement of large language models (LLMs) has made machine-generated text increasingly difficult to distinguish from human-written text. While recent studies explore leveraging internal representations of language models to uncover deeper detection signals, these raw features often exhibit substantial overlap between classes, limiting their discriminative power. To address this challenge, we propose Steer-to-Detect (\texttt{S2D}), a two-stage framework for detecting LLM-generated text. In the first stage, \texttt{S2D} learns a steering vector that is injected into the hidden states of a frozen observer LLM, producing representations with improved class separability. In the second stage, detection is performed via a hypothesis testing procedure based on the steered representations. We establish finite-sample, high-probability guarantees for Type I and Type II errors, providing a theoretical characterization of the procedure. Empirically, \texttt{S2D} achieves strong and consistent performance across a range of settings, including out-of-distribution scenarios and adversarial perturbations.

Machine-Generated Text Detection Language Model Evasion Attacks Hidden States

Open paper Ask about it PDF