Dmitrii Krasheninnikov

19 papers A* 3Journal 14Unranked 2
YearRankTypeTitle / Venue / Authors
2026 J jnl
Moral Preferences of LLMs Under Directed Contextual Influence.
CoRR
Phil Blandfort, Tushar Karayil, Urja Pawar, Robert Graham, Alex McKenzie, Dmitrii Krasheninnikov
2025 J jnl
Detecting High-Stakes Interactions with Activation Probes.
CoRR
Alex McKenzie, Urja Pawar, Phil Blandfort, William Bankes, David Krueger, Ekdeep Singh Lubana, Dmitrii Krasheninnikov
2025 J jnl
Fresh in memory: Training-order recency is linearly encoded in language model activations.
CoRR
Dmitrii Krasheninnikov, Richard E. Turner, David Krueger
2025 J jnl
The Impact of Off-Policy Training Data on Probe Generalisation.
CoRR
Nathalie Kirch, Samuel Dower, Adrians Skapars, Ekdeep Singh Lubana, Dmitrii Krasheninnikov
2025 J jnl
Understanding (Un)Reliability of Steering Vectors in Language Models.
CoRR
Joschka Braun, Carsten Eickhoff, David Krueger, Seyed Ali Bahrainian, Dmitrii Krasheninnikov
2024 J jnl
Comparing Bottom-Up and Top-Down Steering Approaches on In-Context Learning Tasks.
CoRR
Madeline Brumley, Joe Kwon, David Krueger, Dmitrii Krasheninnikov, Usman Anwar
2024 A* conf
Implicit meta-learning may lead language models to trust more reliable sources.
ICML
Dmitrii Krasheninnikov, Egor Krasheninnikov, Bruno Kacper Mlodozeniec, Tegan Maharaj, David Krueger
2024 A* conf
Stress-Testing Capability Elicitation With Password-Locked Models.
NeurIPS
Ryan Greenblatt, Fabien Roger, Dmitrii Krasheninnikov, David Krueger
2024 J jnl
Stress-Testing Capability Elicitation With Password-Locked Models.
CoRR
Ryan Greenblatt, Fabien Roger, Dmitrii Krasheninnikov, David Krueger
2023 — conf
Harms from Increasingly Agentic Algorithmic Systems.
FAccT
Alan Chan, Rebecca Salganik, Alva Markelius, Chris Pang, Nitarshan Rajkumar, Dmitrii Krasheninnikov, Lauro Langosco, Zhonghao He, Yawen Duan, Micah Carroll, Michelle P. Lin, Alex Mayhew, Katherine M. Collins, Maryam Molamohammadi, John Burden, Wanru Zhao, Shalaleh Rismani, Konstantinos Voudouris, Umang Bhatt, Adrian Weller, David Krueger, Tegan Maharaj
2023 J jnl
Harms from Increasingly Agentic Algorithmic Systems.
CoRR
Alan Chan, Rebecca Salganik, Alva Markelius, Chris Pang, Nitarshan Rajkumar, Dmitrii Krasheninnikov, Lauro Langosco, Zhonghao He, Yawen Duan, Micah Carroll, Michelle P. Lin, Alex Mayhew, Katherine M. Collins, Maryam Molamohammadi, John Burden, Wanru Zhao, Shalaleh Rismani, Konstantinos Voudouris, Umang Bhatt, Adrian Weller, David Krueger, Tegan Maharaj
2023 J jnl
Meta- (out-of-context) learning in neural networks.
CoRR
Dmitrii Krasheninnikov, Egor Krasheninnikov, Bruno Mlodozeniec, David Krueger
2023 J jnl
Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback.
CoRR
Stephen Casper, Xander Davies, Claudia Shi, Thomas Krendl Gilbert, Jérémy Scheurer, Javier Rando, Rachel Freedman, Tomasz Korbak, David Lindner, Pedro Freire, Tony Tong Wang, Samuel Marks, Charbel-Raphaël Ségerie, Micah Carroll, Andi Peng, Phillip J. K. Christoffersen, Mehul Damani, Stewart Slocum, Usman Anwar, Anand Siththaranjan, Max Nadeau, Eric J. Michaud, Jacob Pfau, Dmitrii Krasheninnikov, Xin Chen, Lauro Langosco, Peter Hase, Erdem Biyik, Anca D. Dragan, David Krueger, Dorsa Sadigh, Dylan Hadfield-Menell
2023 J jnl
Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback.
Trans. Mach. Learn. Res.
Stephen Casper, Xander Davies, Claudia Shi, Thomas Krendl Gilbert, Jérémy Scheurer, Javier Rando, Rachel Freedman, Tomasz Korbak, David Lindner, Pedro Freire, Tony Tong Wang, Samuel Marks, Charbel-Raphaël Ségerie, Micah Carroll, Andi Peng, Phillip J. K. Christoffersen, Mehul Damani, Stewart Slocum, Usman Anwar, Anand Siththaranjan, Max Nadeau, Eric J. Michaud, Jacob Pfau, Dmitrii Krasheninnikov, Xin Chen, Lauro Langosco, Peter Hase, Erdem Biyik, Anca D. Dragan, David Krueger, Dorsa Sadigh, Dylan Hadfield-Menell
2022 A* conf
Defining and Characterizing Reward Gaming.
NeurIPS
Joar Skalse, Nikolaus H. R. Howe, Dmitrii Krasheninnikov, David Krueger
2022 J jnl
Defining and Characterizing Reward Hacking.
CoRR
Joar Skalse, Nikolaus H. R. Howe, Dmitrii Krasheninnikov, David Krueger
2021 J jnl
Combining Reward Information from Multiple Sources.
CoRR
Dmitrii Krasheninnikov, Rohin Shah, Herke van Hoof
2019 — conf
Preferences Implicit in the State of the World.
ICLR (Poster)
Rohin Shah, Dmitrii Krasheninnikov, Jordan Alexander, Pieter Abbeel, Anca D. Dragan
2019 J jnl
Preferences Implicit in the State of the World.
CoRR
Rohin Shah, Dmitrii Krasheninnikov, Jordan Alexander, Pieter Abbeel, Anca D. Dragan