Raymond Douglas

Researcher, ACS Research

Joined

September 2023

Names

Raymond Douglas (Preferred)

Emails

****@gmail.com (Confirmed)

,

****@acsresearch.org (Confirmed)

,

****@cs.toronto.edu (Confirmed)

Personal Links

Career & Education History

Researcher

ACS Research (acsresearch.org)

2025 – Present

Undergrad student

University of Oxford (ox.ac.uk)

2018 – 2021

Advisors, Relations & Conflicts

No relations added

Expertise

No areas of expertise listed

Publications

Latent Introspection: Models Can Detect Prior Concept Injections
Theia Pearson-Vogel, Martin Vanek, Raymond Douglas, Jan Kulveit
- Mech Interp Workshop ICML 2026 Virtualposter
- Readers: Everyone
Position: We Must Proactively Address AI Safety Debt
Peter Wallich, Raymond Douglas
- ICLR 2026 AIWILD
- Readers: Everyone
Incoherence in Goal-Conditioned Autoregressive Models
Jacek Karwowski, Raymond Douglas
- AISTATS 2026 Poster
- Readers: Everyone
AI Models Can Provably Hide Arbitrary Capabilities
Andis Draguns, Sumeet Ramesh Motwani, Raymond Douglas, Christian Schroeder de Witt
- ICLR 2026 Conference Withdrawn Submission
- Readers: Everyone
Position: Humanity Faces Existential Risk from Gradual Disempowerment
Jan Kulveit, Raymond Douglas, Nora Ammann, Deger Turan, David Krueger, David Duvenaud
- ICML 2025 Position Paper Track poster
- Readers: Everyone
Mitigating the Influence of Distractor Tasks in LMs with Prior-Aware Decoding
Raymond Douglas, Andis Draguns, Tomáš Gavenčiak
- Submitted to ICLR 2025
- Readers: Everyone
Limitations of Agents Simulated by Predictive Models
Raymond Douglas, Jacek Karwowski, Chan Bae, Andis Draguns, Victoria Krakovna
- LLMAgents @ ICLR 2024 Poster
- Readers: Everyone

Co-Authors