:: Library Catalog

Image de couverture de livre

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Kutasov, Jon, Loughridge, Chloe, Sun, Yuqi, Sleight, Henry, Shlegeris, Buck, Tracy, Tyler, Benton, Joe
Format:	Preprint
Publié:	2025
Sujets:	Artificial Intelligence
Accès en ligne:	https://arxiv.org/abs/2511.02997
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Optimizing AI Agent Attacks With Synthetic Data
par: Loughridge, Chloe, et autres
Publié: (2025)

SHADE-Arena: Evaluating Sabotage and Monitoring in LLM Agents
par: Kutasov, Jonathan, et autres
Publié: (2025)

Games for AI Control: Models of Safety Evaluations of AI Deployment Protocols
par: Griffin, Charlie, et autres
Publié: (2024)

Polysemanticity and Capacity in Neural Networks
par: Scherlis, Adam, et autres
Publié: (2022)

SLEIGHT-Bench: A Benchmark of Evasion Attacks Against Agent Monitors
par: Najt, Elle, et autres
Publié: (2026)

dafny-annotator: AI-Assisted Verification of Dafny Programs
par: Poesia, Gabriel, et autres
Publié: (2024)

Model Spec Midtraining: Improving How Alignment Training Generalizes
par: Li, Chloe, et autres
Publié: (2026)

A sketch of an AI control safety case
par: Korbak, Tomek, et autres
Publié: (2025)

How to evaluate control measures for LLM agents? A trajectory from today to superintelligence
par: Korbak, Tomek, et autres
Publié: (2025)

Auditing Sabotage Bench: A Benchmark for Detecting and Fixing Research Sabotage in ML Codebases
par: Gan, Eric, et autres
Publié: (2026)

Language models are better than humans at next-token prediction
par: Shlegeris, Buck, et autres
Publié: (2022)

Adaptive Deployment of Untrusted LLMs Reduces Distributed Threats
par: Wen, Jiaxin, et autres
Publié: (2024)

Subversion Strategy Eval: Can language models statelessly strategize to subvert control protocols?
par: Mallen, Alex, et autres
Publié: (2024)

BashArena: A Control Setting for Highly Privileged AI Agents
par: Kaufman, Adam, et autres
Publié: (2025)

Factor(U,T): Controlling Untrusted AI by Monitoring their Plans
par: Lip, Edward Lue Chee, et autres
Publié: (2025)

Ctrl-Z: Controlling AI Agents via Resampling
par: Bhatt, Aryan, et autres
Publié: (2025)

Sabotage Evaluations for Frontier Models
par: Benton, Joe, et autres
Publié: (2024)

Safe, Untrusted, "Proof-Carrying" AI Agents: toward the agentic lakehouse
par: Tagliabue, Jacopo, et autres
Publié: (2025)

Efficiently Aligning Language Models with Online Natural Language Feedback
par: Ye, Christine, et autres
Publié: (2026)

Singular Vectors of Attention Heads Align with Features
par: Franco, Gabriel, et autres
Publié: (2026)

All Code, No Thought: Current Language Models Struggle to Reason in Ciphered Language
par: Guo, Shiyuan, et autres
Publié: (2025)

The LLM Has Left The Chat: Evidence of Bail Preferences in Large Language Models
par: Ensign, Danielle, et autres
Publié: (2025)

AI Organizations are More Effective but Less Aligned than Individual Agents
par: Shen, Judy Hanwen, et autres
Publié: (2026)

The Hot Mess of AI: How Does Misalignment Scale With Model Intelligence and Task Complexity?
par: Hägele, Alexander, et autres
Publié: (2026)

Attack Selection Reduces Safety in Concentrated AI Control Settings against Trusted Monitoring
par: Schaeffer, Joachim, et autres
Publié: (2026)

Structured Security Auditing and Robustness Enhancement for Untrusted Agent Skills
par: Lv, Lijia, et autres
Publié: (2026)

LinuxArena: A Control Setting for AI Agents in Live Production Software Environments
par: Tracy, Tyler, et autres
Publié: (2026)

Stress-Testing Model Specs Reveals Character Differences among Language Models
par: Zhang, Jifan, et autres
Publié: (2025)

MonitoringBench: Semi-Automated Red-Teaming for Agent Monitoring
par: Jotautaitė, Monika, et autres
Publié: (2026)

A Survey of AI Agent Protocols
par: Yang, Yingxuan, et autres
Publié: (2025)

Inverse Scaling in Test-Time Compute
par: Gema, Aryo Pradipta, et autres
Publié: (2025)

Agent Control Protocol: Admission Control for Agent Actions
par: Fernandez, Marcelo
Publié: (2026)

Bypassing AI Control Protocols via Agent-as-a-Proxy Attacks
par: Isbarov, Jafar, et autres
Publié: (2026)

Data-Centric Interpretability for LLM-based Multi-Agent Reinforcement Learning
par: Yan, John, et autres
Publié: (2026)

PragLocker: Protecting Agent Intellectual Property in Untrusted Deployments via Non-Portable Prompts
par: Li, Qinfeng, et autres
Publié: (2026)

AI Control: Improving Safety Despite Intentional Subversion
par: Greenblatt, Ryan, et autres
Publié: (2023)

Teaching Models to Verbalize Reward Hacking in Chain-of-Thought Reasoning
par: Turpin, Miles, et autres
Publié: (2025)

Removing Sandbagging in LLMs by Training with Weak Supervision
par: Ryd, Emil, et autres
Publié: (2026)

Generative AI in clinical practice: novel qualitative evidence of risk and responsible use of Google's NotebookLM
par: Reuter, Max, et autres
Publié: (2025)

Deep Research Bench: Evaluating AI Web Research Agents
par: FutureSearch, et autres
Publié: (2025)