Aniket Rege

PhD Student, University

of Wisconsin-Madison

Hello There! I’m a Ph.D student in Computer Sciences at UW-Madison, fortunate to be jointly advised by Ramya Vinayak and Yong Jae Lee.

🚨 I’m on the market for research internships in Spring/Summer 2027, focused on agents, RL post-training, and multimodal reasoning!

These days I spend most of my time thinking about:

Long-horizon multimodal reasoning agents - how do we train agents to plan, search, and reflect on their decisions consistently over long reasoning chains? What kind of datasets and evals are needed?
Heterogeneous and diverse preference learning - humans are extremely diverse, and generative models are often trained not to be. How can we do better at catering to such pluralistic preferences?
Broad Interests: elastic embedding models, multi-agent systems, information retrieval, efficient+deployable ML.

Previously, I was a masters student at the University of Washington, Seattle advised by Prof. Ali Farhadi and mentored by Aditya Kusupati. My MS research focused on large-scale elastic and adaptive representation learning (Matryoshka 🪆) and web-scale search.

I was also fortunate to spend time in the industry as a research intern at Google DeepMind (with Gagan Madan), Meta (with Hyo Jin Kim and Yuning Chai), and NVIDIA (with Nikolai Smolyanskiy and Stan Birchfield), and three years as a research engineer at Samsung Research.

If you are an undergrad/masters student interested in working with me, send me an email (I’ll do my best to reply)!

news

Aug 5, 2026	🔥 Playing with Fire is accepted as a spotlight at COLM Efficient Reasoning workshop!
Apr 30, 2026	🎆 Sparks of Cooperative Reasoning is accepted at ICML 2026!
Apr 6, 2026	EGAgent from my internship at Meta Reality Labs is accepted at ACL ‘26 Main Conference!
May 10, 2025	⚕️CuRe is accepted as an oral at CVPR DemoDiv workshop! Update (06/25): CuRe is accepted at ICCV 2025!
Jun 18, 2024	PAL is accepted at ICML workshops: TF2M and MHFAIA (oral)! Update (01/22/2025): PAL is accepted at ICLR 2025!

selected publications

ICML

Sparks of Cooperative Reasoning: LLMs as Strategic Hanabi Agents

Mahesh Ramesh, Kaousheik Jayakumar, Aswinkumar Ramkumar, Pavan Thodima, Aniket Rege†, and Manolis Vlatakis†

In Forty-third International Conference on Machine Learning (ICML), 2026

arXiv Bib Blog Code

Citations 6

@inproceedings{ramesh2026sparks,
  title = {Sparks of Cooperative Reasoning: {LLM}s as Strategic Hanabi Agents},
  author = {Ramesh, Mahesh and Jayakumar, Kaousheik and Ramkumar, Aswinkumar and Thodima, Pavan and Rege†, Aniket and Vlatakis†, Manolis},
  booktitle = {Forty-third International Conference on Machine Learning (ICML)},
  year = {2026},
}

ACL

Agentic Very Long Video Understanding

Aniket Rege, Arka Sadhu, Yuliang Li, Kejie Li, Ramya Korlakai Vinayak, Yuning Chai, Yong Jae Lee, and Hyo Jin Kim

In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (ACL), 2026

🚨 Recommended Spotlight

arXiv Bib Blog Code

Citations 20

@inproceedings{rege2025agentic,
  title = {Agentic Very Long Video Understanding},
  author = {Rege, Aniket and Sadhu, Arka and Li, Yuliang and Li, Kejie and Korlakai Vinayak, Ramya and Chai, Yuning and Lee, Yong Jae and Kim, Hyo Jin},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (ACL)},
  year = {2026},
}

ICLR

PAL: Sample-Efficient Personalized Reward Modeling for Pluralistic Alignment

Daiwei Chen, Yi Chen, Aniket Rege, Zhi Wang, and Ramya Korlakai Vinayak

In The Thirteenth International Conference on Learning Representations (ICLR), 2025

arXiv Bib Blog Code

Citations 121

@inproceedings{chen2025pal,
  title = {PAL: Sample-Efficient Personalized Reward Modeling for Pluralistic Alignment},
  author = {Chen, Daiwei and Chen, Yi and Rege, Aniket and Wang, Zhi and Korlakai Vinayak, Ramya},
  booktitle = {The Thirteenth International Conference on Learning Representations (ICLR)},
  year = {2025},
}

NeurIPS

Matryoshka Representation Learning

Aditya Kusupati*, Gantavya Bhatt*, Aniket Rege*, Matthew Wallingford, Aditya Sinha, Vivek Ramanujan, William Howard-Snyder, Kaifeng Chen, Sham Kakade, Prateek Jain, and Ali Farhadi

In Advances in Neural Information Processing Systems (NeurIPS), 2022

arXiv Bib Blog Code Talk

Citations 720

@inproceedings{kusupati2022matryoshka,
  title = {Matryoshka Representation Learning},
  author = {Kusupati*, Aditya and Bhatt*, Gantavya and Rege*, Aniket and Wallingford, Matthew and Sinha, Aditya and Ramanujan, Vivek and Howard-Snyder, William and Chen, Kaifeng and Kakade, Sham and Jain, Prateek and Farhadi, Ali},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2022},
}