Buck Shlegeris

4 papers · 2022–2025 · 3 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓

🌍 Conference Polyglot · Bronze 🌉 Interdisciplinary Bridge · Bronze 🗺️ Taxonomy Completionist (10) 🐣 Hot Topic Early Bird

Venues

ICLR (2) ICML (1) NeurIPS (1)

Top co-authors

Ethan Perez (1) Adam Scherlis (1) Peter Schmidt-Nielsen (1) Henry Sleight (1) Fabien Roger (1) Caleb Larson (1) Shi Feng (1) Lawrence Chan (1) Noa Nabeshima (1) Kevin Ro Wang (1)

Research topics

Artificial Intelligence (4) Core AI (4) Deep Learning (2) Machine Learning (2) Natural Language Processing (1) Application Areas (1) Generation (1) Learning Types · Machine Learning (1) Learning Types · Deep Learning (1) Optimization & Theory (1)

Keywords

large language model (2) text generation (1) adversarial training (1) ai safety (1) worst-case performance (1) language model (1) safety evaluation (1) mechanistic interpretability (1) attention head (1) model deployment (1) causal intervention (1) backdoor detection (1) reverse engineering (1) classifier threshold (1) high-stakes reliability (1) indirect object identification (1) editing code (1) monitoring model (1) subversion (1) trusted model (1)

Papers

Adaptive Deployment of Untrusted LLMs Reduces Distributed Threats ICLR 2025 AI Control: Improving Safety Despite Intentional Subversion ICML 2024 Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 Small ICLR 2023 Adversarial training for high-stakes reliability NeurIPS 2022