Alexander Matt Turner

21 papers A* 3C 1Journal 15Unranked 2

Year	Rank	Type	Title / Venue / Authors
2025	J	`jnl`	An Approach to Technical AGI Safety and Security. CoRR Rohin Shah, Alex Irpan, Alexander Matt Turner, Anna Wang, Arthur Conmy, David Lindner, Jonah Brown-Cohen, Lewis Ho, Neel Nanda, Raluca Ada Popa, Rishub Jain, Rory Greig, Samuel Albanie, Scott Emmons, Sebastian Farquhar, Sébastien Krier, Senthooran Rajamanoharan, Sophie Bridgers, Tobi Ijitoye, Tom Everitt, Victoria Krakovna, Vikrant Varma, Vladimir Mikulik, Zachary Kenton, Dave Orr, Shane Legg, Noah D. Goodman, Allan Dafoe, Four Flynn, Anca D. Dragan
2025	J	`jnl`	Consistency Training Helps Stop Sycophancy and Jailbreaks. CoRR Alex Irpan, Alexander Matt Turner, Mark Kurzeja, David K. Elson, Rohin Shah
2025	J	`jnl`	Distillation Robustifies Unlearning. CoRR Bruce W. Lee, Addie Foote, Alex Infanger, Leni Shor, Harish Kamath, Jacob Goldman-Wetzler, Bryce Woodworth, Alex Cloud, Alexander Matt Turner
2025	J	`jnl`	Output Supervision Can Obfuscate the Chain of Thought. CoRR Jacob Drori, Luke Marks, Bryce Woodworth, Alex Cloud, Alexander Matt Turner
2025	J	`jnl`	Recontextualization Mitigates Specification Gaming without Modifying the Specification. CoRR Ariana Azarbal, Victor Gillioz, Vladimir Ivanov, Bryce Woodworth, Jacob Drori, Nevan Wichers, Aram Ebtekar, Alex Cloud, Alexander Matt Turner
2024	J	`jnl`	Gradient Routing: Masking Gradients to Localize Computation in Neural Networks. CoRR Alex Cloud, Jacob Goldman-Wetzler, Evzen Wybitul, Joseph Miller, Alexander Matt Turner
2024	—	`conf`	Steering Llama 2 via Contrastive Activation Addition. ACL (1) Nina Rimsky, Nick Gabrieli, Julian Schulz, Meg Tong, Evan Hubinger, Alexander Matt Turner
2023	J	`jnl`	Activation Addition: Steering Language Models Without Optimization. CoRR Alexander Matt Turner, Lisa Thiergart, David Udell, Gavin Leech, Ulisse Mini, Monte MacDiarmid
2023	J	`jnl`	Steering Llama 2 via Contrastive Activation Addition. CoRR Nina Rimsky, Nick Gabrieli, Julian Schulz, Meg Tong, Evan Hubinger, Alexander Matt Turner
2023	J	`jnl`	Understanding and Controlling a Maze-Solving Policy Network. CoRR Ulisse Mini, Peli Grietzer, Mrinank Sharma, Austin Meek, Monte MacDiarmid, Alexander Matt Turner
2022	J	`jnl`	Formalizing the Problem of Side Effect Regularization. CoRR Alexander Matt Turner, Aseem Saxena, Prasad Tadepalli
2022	J	`jnl`	On Avoiding Power-Seeking by Artificial Intelligence. CoRR Alexander Matt Turner
2022	A*	`conf`	Parametrically Retargetable Decision-Makers Tend To Seek Power. NeurIPS Alexander Matt Turner, Prasad Tadepalli
2022	J	`jnl`	Parametrically Retargetable Decision-Makers Tend To Seek Power. CoRR Alexander Matt Turner, Prasad Tadepalli
2021	A*	`conf`	Optimal Policies Tend To Seek Power. NeurIPS Alexander Matt Turner, Logan Smith, Rohin Shah, Andrew Critch, Prasad Tadepalli
2020	A*	`conf`	Avoiding Side Effects in Complex Environments. NeurIPS Alexander Matt Turner, Neale Ratzlaff, Prasad Tadepalli
2020	J	`jnl`	Avoiding Side Effects in Complex Environments. CoRR Alexander Matt Turner, Neale Ratzlaff, Prasad Tadepalli
2020	C	`conf`	Conservative Agency via Attainable Utility Preservation. AIES Alexander Matt Turner, Dylan Hadfield-Menell, Prasad Tadepalli
2019	J	`jnl`	Conservative Agency via Attainable Utility Preservation. CoRR Alexander Matt Turner, Dylan Hadfield-Menell, Prasad Tadepalli
2019	—	`conf`	Conservative Agency. AISafety@IJCAI Alexander Matt Turner, Dylan Hadfield-Menell, Prasad Tadepalli
2019	J	`jnl`	Optimal Farsighted Agents Tend to Seek Power. CoRR Alexander Matt Turner