Emanuel Tewolde
I am a fourth-year PhD student in the Computer Science Department of Carnegie Mellon University, where I am fortunate enough to be advised by Vincent Conitzer. My work is supported in part by the Cooperative AI PhD Fellowship.
I strive to understand how to enable artificial intelligence and humans to effectively achieve better (social) outcomes in strategic interactions with other agents. More specifically, my current research interests lie in algorithmic game theory, LLM agents, and reinforcement learning, with an emphasis on the safety, coordination, cooperation, and alignment of AI systems. Further research topics I enjoy are mathematical optimization, learning in games, computational complexity, and social choice theory.
Currently, I am working at Meta’s research unit FAIR (within Meta Superintelligence Labs) on LLM coding agents for AI research. Prior to CMU, I completed a master’s and bachelor’s degree in mathematics at Imperial College London and the Technical University Darmstadt respectively. In addition to that, I have also worked with the Fraunhofer-Gesellschaft (IEE) on machine learning methods for smarter renewable energy systems.
Feel free to reach out to me under emanueltewolde (at) cmu (dot) edu.
CV, Google Scholar, DBLP
News
- • Excited to join FAIR at Meta as a research engineer to work on LLM coding agents for AI research, together with Yoram Bachrach, Jakob Foerster, and others! Aug 2025
- • Visiting the University of Oxford again for the summer! May 2025
- • One paper accepted to ICML 2025 as an oral and one paper accepted to EC 2025! Apr 2025
- • Very honored to be awarded the AAAI 2025 Best Poster Award! (out of 674 posters) Mar 2025
- • Very honored to be awarded the Cooperative AI PhD Fellowship starting from 2025! Dec 2024
- • Two papers accepted to AAAI 2025! Reach out if you are going to be there :) Dec 2024
- • Visited the Algorithmic Game Theory Group at Google DeepMind in London to present our work on Social Choice for AI Alignment; thanks for having me! May 2024
- • Visiting the University of Oxford again for the summer! May 2024
- • Two papers accepted to IJCAI 2024 and one paper accepted at ICML 2024! Reach out if you are going to be there :) Apr 2024
- • Visiting the University of Oxford again for the summer! May 2023
- • My first academic paper got accepted to IJCAI 2023! Reach out if you are going to be there :) Apr 2023
- • Excited to start my PhD at Carnegie Mellon University in the group of Prof. Vincent Conitzer! Aug 2022
- • Visiting the University of Oxford for the summer to work with Prof. Vincent Conitzer. Thank you to the Cooperative AI Foundation for funding my research visit! May 2022
- • Personal Website launched! (Though it is still a work in progress...) Oct 2021
Papers
In below, ‘ - αβ - |’ stands for alphabetical author ordering, and ‘==’ superscripts stand for equal contribution.
Working Papers
Decision Making under Imperfect Recall: Algorithms and Benchmarks
Emanuel Tewolde, Brian Hu Zhang, Ioannis Anagnostides, Tuomas Sandholm, Vincent ConitzerOn the Edge of Core (Non-)Emptiness: An Automated Reasoning Approach to Approval-Based Multi-Winner Voting
Ratip Emin Berker==, Emanuel Tewolde==, Vincent Conitzer, Mingyu Guo, Marijn Heule, Lirong XiaWhen Ethics and Payoffs Diverge: LLM Agents in Morally Charged Social Dilemmas
Steffen Backmann, David Guzman Piedrahita, Emanuel Tewolde, Rada Mihalcea, Bernhard Schölkopf, Zhijing Jin[arXiv]
Publications
Evaluating Generalization Capabilities of LLM-Based Agents in Mixed-Motive Scenarios Using Concordia
Chandler Smith, Marwa Abdulhai, ... (15 more authors) ..., Emanuel Tewolde, ... (66 more authors) ..., Jose Hernandez-Orallo, Joel Z LeiboTo appear in Neural Information Processing Systems (NeurIPS) 2025 -- Datasets and Benchmarks Track
[published version]
Expected Variational Inequalities
Brian Hu Zhang==, Ioannis Anagnostides==, Emanuel Tewolde, Ratip Emin Berker, Gabriele Farina, Vincent Conitzer, Tuomas SandholmPublished in International Conference on Machine Learning (ICML) 2025
Oral (Top 1.0% of submissions)
[published version] [arXiv]
Learning and Computation of Φ-Equilibria at the Frontier of Tractability
Brian Hu Zhang==, Ioannis Anagnostides==, Emanuel Tewolde, Ratip Emin Berker, Gabriele Farina, Vincent Conitzer, Tuomas SandholmPublished in ACM Conference on Economics and Computation (EC) 2025
[published version] [arXiv]
Computing Game Symmetries and Equilibria That Respect Them
Emanuel Tewolde, Brian Hu Zhang, Caspar Oesterheld, Tuomas Sandholm, and Vincent ConitzerPublished in Association for the Advancement of Artificial Intelligence (AAAI) 2025
Oral (Top 4.6% of submissions)
Best Poster Award (out of 674 posters)
[published version] [arXiv] [video]
The Value of Recall in Extensive-Form Games
Ratip Emin Berker, Emanuel Tewolde, Ioannis Anagnostides, Tuomas Sandholm, and Vincent ConitzerPublished in Association for the Advancement of Artificial Intelligence (AAAI) 2025
Oral (Top 4.6%)
[published version] [arXiv]
Imperfect-Recall Games: Equilibrium Concepts and Their Complexity
Emanuel Tewolde, Brian Hu Zhang, Caspar Oesterheld, Manolis Zampetakis, Tuomas Sandholm, Paul W. Goldberg, and Vincent ConitzerPublished in International Joint Conference on Artificial Intelligence (IJCAI) 2024
[published version] [arXiv]
Game Transformations That Preserve Nash Equilibria or Best-Response Sets
Emanuel Tewolde and Vincent ConitzerPublished in International Joint Conference on Artificial Intelligence (IJCAI) 2024
[published version] [arXiv]
Social Choice Should Guide AI Alignment in Dealing with Diverse Human Feedback
- αβ - | Vincent Conitzer, Rachel Freedman, Jobst Heitzig, Wesley H. Holliday, Bob M. Jacobs, Nathan Lambert, Milan Mossé, Eric Pacuit, Stuart Russell, Hailey Schoelkopf, Emanuel Tewolde, and William S. ZwickerPublished in International Conference on Machine Learning (ICML) 2024 -- Position Paper Track
[published version] [arXiv] [video] [Featured on Interconnects]
The Computational Complexity of Single-Player Imperfect-Recall Games
Emanuel Tewolde, Caspar Oesterheld, Vincent Conitzer, and Paul W. GoldbergPublished in International Joint Conference on Artificial Intelligence (IJCAI) 2023
[published version] [arXiv] [video]
Teaching
Taught and supervised 5 – 10 new TAs per semester for the mathematics department