Sam Toyer

OpenAI (United States) — United States

Computer Science · Physical Sciences

9h-index400citations20works9.52yr avg
OpenAlex↗ORCID↗Google Scholar↗Lab page↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Reinforcement Learning in Robotics(5), Adversarial Robustness in Machine Learning(5), AI-based Problem Solving and Planning(3), Machine Learning and Algorithms(3), Natural Language Processing Techniques(2)

Publications19 total

Deliberative Alignment: Reasoning Enables Safer Language Models
SuperIntelligence - Robotics - Safety & Alignment·2025· 12 citedOpen Access
Trading Inference-Time Compute for Adversarial Robustness
arXiv (Cornell University)·2025Open Access
Deliberative Alignment: Reasoning Enables Safer Language Models
arXiv (Cornell University)·2024· 9 citedOpen Access
A StrongREJECT for Empty Jailbreaks
arXiv (Cornell University)·2024· 4 citedOpen Access
A StrongREJECT for Empty Jailbreaks
2024
Tensor Trust: Interpretable Prompt Injection Attacks from an Online Game
arXiv (Cornell University)·2023· 5 citedOpen Access
imitation: Clean Imitation Learning Implementations
arXiv (Cornell University)·2022· 9 citedOpen Access
An Empirical Investigation of Representation Learning for Imitation
arXiv (Cornell University)·2022· 7 citedOpen Access
A Primer on Maximum Causal Entropy Inverse Reinforcement Learning
arXiv (Cornell University)·2022· 2 citedOpen Access
Guiding Search with Generalized Policies for Probabilistic Planning
Proceedings of the International Symposium on Combinatorial Search·2021· 12 citedOpen Access
DERAIL: Diagnostic Environments for Reward And Imitation Learning
arXiv (Cornell University)·2020· 1 citedOpen Access
The MAGICAL Benchmark for Robust Imitation
arXiv (Cornell University)·2020· 7 citedOpen Access
The MAGICAL Benchmark for Robust Imitation
arXiv (Cornell University)·2020· 1 citedOpen Access
Variational Discriminator Bottleneck: Improving Imitation Learning, Inverse RL, and GANs by Constraining Information Flow
arXiv (Cornell University)·2018· 101 citedOpen Access
Variational Discriminator Bottleneck: Improving Imitation Learning, Inverse RL, and GANs by Constraining Information Flow
arXiv (Cornell University)·2018· 73 citedOpen Access
Action Schema Networks: Generalised Policies With Deep Learning
Proceedings of the AAAI Conference on Artificial Intelligence·2018· 55 citedOpen Access
Human Pose Forecasting via Deep Markov Models
2017· 41 cited
Action Schema Networks: Generalised Policies with Deep Learning
arXiv (Cornell University)·2017· 27 citedOpen Access
Human Pose Forecasting via Deep Markov Models
arXiv (Cornell University)·2017· 2 citedOpen Access

Frequent Co-authors

Pieter Abbeel(6), Stuart Russell(5), Scott Emmons(4), Felipe Trevizan(3), Sylvie Thiébaux(3), Lexing Xie(3), Roger D. Dias(3), Amelia Glaese(3), Adam Gleave(3), Rohin Shah(3), Olivia Watkins(3), Justin Svegliato(3), Xue Bin Peng(2), Angjoo Kanazawa(2), Sergey Levine(2), Anoop Cherian(2), Tengda Han(2), Stephen Jay Gould(2), Melody Y. Guan(2), Manas Joglekar(2)