I am an Associate Professor of EECS at MIT, where I run the Algorithmic Alignment Group in the Computer Science and Artificial Intelligence Laboratory (CSAIL). I work on AI alignment: developing methods to ensure that AI systems behave in accordance with the goals and values of their users and of society as a whole. My group's research spans interpreting and evaluating the internals of modern AI systems, learning robustly from human feedback, and designing the multi-agent and societal structures needed to oversee them. Our goal is to enable the safe, beneficial, and trustworthy deployment of AI in real-world settings.
Dylan Hadfield-Menell
dhm at csail.mit.edu
Associate Professor of EECS
Faculty of Artificial
Intelligence and Decision-Making
Computer Science and
Artificial Intelligence Laboratory
Dept. of Electrical
Engineering and Computer Science
Massachusetts
Institute of Technology
44 Vassar St., 45-701D
Cambridge, MA 02139
United States
Publications
Journal Publications- Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities. Zora Che, Stephen Casper, Robert Kirk, Anirudh Satheesh, Stewart Slocum, Lev McKinney, Rohit Gandikota, Aidan Ewart, Domenic Rosati, Zichu Wu, Zikui Cai, Bilal Chughtai, Yarin Gal, Furong Huang, Dylan Hadfield-Menell. Transactions on Machine Learning Research (TMLR). 2025.
- Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMs Abhay Sheshadri, Aidan Ewart, Phillip Guo, Aengus Lynch, Cindy Wu, Vivek Hebbar, Henry Sleight, Asa Cooper Stickland, Ethan Perez, Dylan Hadfield-Menell, Stephen Casper. Transactions on Machine Learning Research (TMLR). 2025.
- Defending Against Unforeseen Failure Modes with Latent Adversarial Training Stephen Casper, Lennart Schulze, Oam Patel, Dylan Hadfield-Menell. Transactions on Machine Learning Research (TMLR). 2025.
- Building Human Values into Recommender Systems: An Interdisciplinary Synthesis Jonathan Stray, Alon Halevy, Parisa Assar, Dylan Hadfield-Menell, Craig Boutilier, Amar Ashar, Chloe Bakalar, Lex Beattie, Michael Ekstrand, Claire Leibowicz, Connie Moon Sehat, Sara Johansen, Lianne Kerlin, David Vickrey, Spandana Singh, Sanne Vrijenhoek, Amy Zhang, McKane Andrus, Natali Helberger, Polina Proutskova, Tanushree Mitra, Nina Vasan. ACM Transactions on Recommender Systems. 2024.
- Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback Stephen Casper, Xander Davies, Claudia Shi, Thomas Krendl Gilbert, Jérémy Scheurer, Javier Rando, Rachel Freedman, Tomasz Korbak, David Lindner, Pedro Freire, Tony Wang, Samuel Marks, Charbel-Raphaël Segerie, Micah Carroll, Andi Peng, Phillip 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 Bıyık, Anca Dragan, David Krueger, Dorsa Sadigh, Dylan Hadfield-Menell. Transactions on Machine Learning Research (TMLR). 2023.
- Judging Facts, Judging Norms: Training Machine Learning Models to Judge Humans Requires a Modified Approach to Labeling Data Aparna Balagopalan, David Madras, David H Yang, Dylan Hadfield-Menell, Gillian K Hadfield, Marzyeh Ghassemi. Science Advances. 2023.
- Spurious Normativity Enhances Learning of Compliance and Enforcement Behavior in Artificial Agents. Raphael Koster, Dylan Hadfield-Menell, Richard Everett, Laura Weidinger, Gillian K. Hadfield, and Joel Z. Leibo. Proceedings of the National Academy of Sciences of the United States (PNAS). 2022.
- When Curation Becomes Creation: Algorithms, Microcontent, and the Vanishing Distinction Between Platforms and Creators. Liu Leqi, Dylan Hadfield-Menell, Zachary C. Lipton. Communications of the ACM. 2021.
- Randomness, Not Representation: The Unreliability of Evaluating Cultural Alignment in LLMs. Ariba Khan, Stephen Casper, Dylan Hadfield-Menell. Proceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency (FAccT). 2025.
- Diverse Preference Learning for Capabilities and Alignment. Stewart Slocum, Asher Parker-Sartori, Dylan Hadfield-Menell. The Thirteenth International Conference on Learning Representations (ICLR). 2025.
- Pitfalls of Evidence-Based AI Policy. Stephen Casper, David Krueger, Dylan Hadfield-Menell. ICLR Blogpost Track. 2025.
- Melting Pot Contest: Charting the Future of Generalized Cooperative Intelligence. Rakshit Trivedi, Akbir Khan, Jesse Clifton, Lewis Hammond, Edgar A Duéñez-Guzmán, John P Agapiou, Jayd Matyas, Sasha Vezhnevets, Dipam Chakraborty, Yue Zhao, Marko Tesic, Barna Pásztor, Yunke Ao, Omar G. Younis, Jiawei Huang, Benjamin Swain, Haoyuan Qin, Mian Deng, Ziwei Deng, Utku Erdoganaras, Natasha Jaques, Jakob Nicolaus Foerster, Vincent Conitzer, Jose Hernandez-Orallo, Dylan Hadfield-Menell, Joel Z Leibo. Advances in Neural Information Processing Systems (NeurIPS), Competition Track. 2024.
- The Concordia Contest: Advancing the Cooperative Intelligence of Language Agents. Chandler Smith, Rakshit Trivedi, Jesse Clifton, Lewis Hammond, Akbir Khan, Marwa Abdulhai, Alexander Sasha Vezhnevets, John P Agapiou, Edgar A Duéñez-Guzmán, Jayd Matyas, Danny Karmon, Oliver Slumbers, Minsuk Chang, Dylan Hadfield-Menell, Natasha Jaques, Tim Baarslag, Joel Z. Leibo. Advances in Neural Information Processing Systems (NeurIPS), Competition Track. 2024.
- Black-Box Access is Insufficient for Rigorous AI Audits. Stephen Casper, Carson Ezell, Charlotte Siegmann, Noam Kolt, Taylor Lynn Curtis, Benjamin Bucknall, Andreas Haupt, Kevin Wei, Jérémy Scheurer, Marius Hobbhahn, Lee Sharkey, Satyapriya Krishna, Marvin Von Hagen, Silas Alberti, Alan Chan, Qinyi Sun, Michael Gerovitch, David Bau, Max Tegmark, David Krueger, Dylan Hadfield-Menell. Proceedings of the 2024 ACM Conference on Fairness, Accountability, and Transparency (FAccT). 2024.
- Distributional Preference Learning: Understanding and Accounting for Hidden Context in RLHF. Anand Siththaranjan, Cassidy Laidlaw, Dylan Hadfield-Menell. Proceedings of the 12th International Conference on Learning Representations (ICLR). 2024.
- The SaTML 24 CNN Interpretability Competition: New Innovations for Concept-Level Interpretability. Stephen Casper, Jieun Yun, Joonhyuk Baek, Yeseong Jung, Minhwan Kim, Kiwan Kwon, Saerom Park, Hayden Moore, David Shriver, Marissa Connor, Keltin Grimes, Angus Nicolson, Arush Tagade, Jessica Rumbelow, Hieu Minh Nguyen, Dylan Hadfield-Menell. roceedings of the 2024 IEEE Conference on Secure and Trustworthy Machine Learning (SaTML), 2024.
- Red Teaming AI: The Devil Is In The Details. Sina Fazelpour, Dylan Hadfield-Menell, Luca Belli. Tech Policy Press. 2024.
- Cognitive Dissonance: Why Do Language Model Outputs Disagree with Internal Representations of Truthfulness? Kevin Liu, Stephen Casper, Dylan Hadfield-Menell, Jacob Andreas. Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP). 2023.
- Get It in Writing: Formal Contracts Mitigate Social Dilemmas in Multi-Agent RL. Phillip JK Christoffersen, Andreas A Haupt, Dylan Hadfield-Menell. Proceedings of the 2023 International Conference on Autonomous Agents and Multiagent Systems (AAMAS). 2023.
- Red Teaming Deep Neural Networks with Feature Synthesis Tools. Stephen Casper, Yuxiao Li, Jiawei Li, Tong Bu, Kevin Zhang, Kaivalya Hariharan, Dylan Hadfield-Menell. Advances in Neural Information Processing Systems (NeurIPS). 2023.
- Toward Transparent AI: A Survey on Interpreting the Inner Structures of Deep Neural Networks. Tilman Räuker, Anson Ho, Stephen Casper, Dylan Hadfield-Menell. 2023 IEEE Conference on Secure and Trustworthy Machine Learning (SaTML). 2023.
- Estimating and Penalizing Induced Preference Shifts in Recommender Systems. Micah D Carroll, Anca Dragan, Stuart Russell, Dylan Hadfield-Menell. International Conference on Machine Learning (ICML). 2022.
- A Penalty Default Approach to Preemptive Harm Disclosure and Mitigation for AI Systems. Rui-Jie Yew, Dylan Hadfield-Menell. Proceedings of the 2022 AAAI/ACM Conference on AI, Ethics, and Society (AIES). 2022.
- Robust Feature-Level Adversaries are Interpretability Tools. Stephen Casper, Max Nadeau, Dylan Hadfield-Menell, Gabriel Kreiman. Advances in Neural Information Processing Systems (NeurIPS). 2022.
- How to Talk so AI will Learn: Instructions, Descriptions, and Autonomy. Theodore R Sumers, Robert D Hawkins, Mark K Ho, Thomas L Griffiths, Dylan Hadfield-Menell. Advances in Neural Information Processing Systems (NeurIPS). 2022.
- Towards Psychologically-Grounded Dynamic Preference Models. Mihaela Curmei, Andreas A Haupt, Dylan Hadfield-Menell, Benjamin Recht. Proceedings of the 16th ACM Conference on Recommender Systems (RecSys). 2022.
- Guided Imitation of Task and Motion Planning. Michael J. McDonald, Dylan Hadfield-Menell. Proceedings of the 5th Conference on Robot Learning (CoRL). 2021.
- Consequences of Misaligned AI. Simon Zhuang, Dylan Hadfield-Menell. Proceedings of the 34th Conference on Neural Information Processing Systems (NeurIPS). 2020.
- Conservative Agency via Attainable Utility Preservation. Alexander M. Turner, Dylan Hadfield-Menell, Prasad Tadepalli. Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society (AIES). 2020.
- The Assistive Multi-Armed Bandit. Lawrence Chan, Dylan Hadfield-Menell, Siddartha S. Srinivasa, Anca D. Dragan. Proceedings of the ACM/IEEE International Conference on Human-Robot Interaction (HRI). 2019.
- Human-AI Learning Performance in Multi-Armed Bandits. Ravi Pandya, Sandy H. Huang, Dylan Hadfield-Menell, Anca D. Dragan. Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society (AIES). 2019.
- Legible Normativity for AI Alignment: The Value of Silly Rules. Dylan Hadfield-Menell, McKane Andrus, Gillian K. Hadfield. Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society (AIES). 2019.
- Incomplete Contracting and AI Alignment. Dylan Hadfield-Menell, Gillian K. Hadfield. Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society (AIES). 2019.
- An Efficient, Generalized Bellman Update for Cooperative Inverse Reinforcement Learning. Dhruv Malik, Malayandi Palaniappan, Jaime F. Fisac, Dylan Hadfield-Menell, Stuart J. Russell, Anca D. Dragan. Proceedings of the International Conference on Machine Learning Research (ICML). 2018.
- Simplifying Reward Design through Divide-and-Conquer. Ellis Ratner, Dylan Hadfield-Menell, Anca D. Dragan. Proceedings of Robotics: Science and Systems (RSS). 2018.
- Pragmatic-Pedagogic Value Alignment. Jaime F. Fisac, Monica A. Gates, Jessica B. Hamrick, Chang Liu, Dylan Hadfield-Menell, Malayandi Palaniappan, Dhruv Malik, S. Shankar Sastry, Thomas L. Griffiths, Anca D. Dragan. Proceedings of the International Symposium on Robotics Research (ISRR). 2017.
- Inverse Reward Design. Dylan Hadfield-Menell, Smitha Milli, Pieter Abbeel, Stuart J. Russell, Anca D. Dragan. Proceedings of the 31st Conference on Neural Information Processing Systems (NeurIPS). 2017.
- The Off-Switch Game. Dylan Hadfield-Menell, Anca D. Dragan, Pieter Abbeel, Stuart J. Russell. Proceedings of the 26th International Joint Conference on Artificial Intelligence (IJCAI). 2017.
- Should Robots be Obedient? Smitha Milli, Dylan Hadfield-Menell, Anca D. Dragan, Stuart J. Russell. Proceedings of the 26th International Joint Conference on Artificial Intelligence (IJCAI), 2017.
- Expressive Robot Motion Timing. Allan Zhou, Dylan Hadfield-Menell, and Anca D. Dragan. Proceedings of the ACM/IEEE International Conference on Human-Robot Interaction. 2017.
- Cooperative Inverse Reinforcement Learning. Dylan Hadfield-Menell, Anca D. Dragan, Pieter Abbeel, Stuart J. Russell. Proceedings of the 30th Conference on Neural Information Processing Systems (NeurIPS), 2016.
- Sequential Quadratic Programming for Task Plan Optimization. Dylan Hadfield-Menell, Chris Lin, Rohan Chitnis, Pieter Abbeel, Stuart J. Russell. Proceedings of the IEEE/RSJ Conference on Intelligent Robots and Systems (IROS). 2016.
- Guided Search for Task and Motion Plans Using Learned Heuristics. Rohan Chitnis, Dylan Hadfield-Menell, Abhishek Gupta, Siddharth Srivastava, Edward Groshev, Christopher Lin, Pieter Abbeel. Proceedings of the IEEE Conference on Robotics and Automation (ICRA). 2016.
- Modular Task and Motion Planning in Belief Space. Dylan Hadfield-Menell, Edward Groshev, Rohan Chitnis, and Pieter Abbeel. Proceedings of the IEEE/RSJ Conference on Intelligent Robots and Systems (IROS). 2015.
- Multitasking: Efficient Optimal Planning for Bandit Superprocesses. Dylan Hadfield-Menell, Stuart J. Russell. Proceedings of the 31st Conference on Uncertainty in Artificial Intelligence (UAI). 2015. Supplementary Materials.
- Beyond Lowest-Warping Cost Action Selection in Trajectory Transfer. Dylan Hadfield-Menell, Alex X. Lee, Chelsea Finn, Eric Tzeng, Sandy Huang, and Pieter Abbeel. Proceedings of the IEEE Conference on Robotics and Automation (ICRA). 2015.
- Unifying Scene Registration and Trajectory Optimization for Learning from Demonstrations with Application to Manipulation of Deformable Objects. Alex X. Lee, Sandy H. Huang, Dylan Hadfield-Menell, Eric Tzeng, Pieter Abbeel. Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). 2014.
- Optimization in the Now: Dynamic Peephole Optimization for Hierarchical Planning Dylan Hadfield-Menell, Leslie Pack Kaelbling, and Tomás Lozano-Pérez. Proceedings of the IEEE Conference on Robotics and Automation (ICRA). 2013.
- What are you optimizing for? Aligning Recommender Systems with Human Values. Jonathan Stray, Ivan Vendrov, Jeremy Nixon, Steven Adler, Dylan Hadfield-Menell. ICML Workshop on Participatory Approaches to Machine Learning. 2020.
- Multi-Principal Assistance Games: Definition and Collegial Mechanisms. Arnaud Fickinger, Simon Zhuang, Andrew Critch, Dylan Hadfield-Menell, Stuart Russell. NeurIPS Workshop on Cooperative AI. 2020.
- An Extensible Interactive Interface for Agent Design. Matthew Rahtz, James Fang, Anca D. Dragan, Dylan Hadfield-Menell. 2019.
- Adversarial Training with Voronoi Constraints. Marc Khoury, Dylan Hadfield-Menell. 2019.
- The
Principal-Agent Value Alignment Problem in
Artificial Intelligence. Ph.D. Thesis.
Advisors: Anca D. Dragan, Pieter Abbeel, Stuart J. Russell - Execution
Cost Optimization for Hierarchical Planning in the
Now. M.Eng. Thesis.
Advisors: Tomas Lozano-Perez, Leslie Pack Kaelbling
Information for Prospective Ph.D. Students
How to Express Interest
Please do not email me directly. I can't reliably respond in a fair way. Instead, please fill out our prospective student interest form. This help will help me efficiently review all inquiries and ensure I don't miss anyone during the official application review.
Meeting Requests
Due to the high volume of inquiries, I typically cannot meet with prospective students outside the standard admissions process.
Diversity and Inclusion
I am committed to building a diverse and inclusive research group. I strongly encourage applications from students of all backgrounds, particularly those from underrepresented groups in STEM.
Application Process
For detailed information about the application process, deadlines, and requirements, please visit the EECS Graduate Admissions page.