Accepted abstracts

2025

1. Alignment as Distribution Learning: Your Preference Model is Explicitly a Language Model

   Jihun Yun, Juno Kim, Jongho Park, Junhyuck Kim, Jongha Jon Ryu, Jaewoong Cho, and Kwang-Sung Jun

   2025

   Bib

   @article{paper1,
     title = {Alignment as Distribution Learning: Your Preference Model is Explicitly a Language Model},
     author = {Yun, Jihun and Kim, Juno and Park, Jongho and Kim, Junhyuck and Jon Ryu, Jongha and Cho, Jaewoong and Jun, Kwang-Sung},
     year = {2025}
   }

2. Second-Order Bounds for [0,1]-Valued Regression via Betting Loss

   Yinan Li, Ethan Huang, Sungjoon Yoon, and Kwang-Sung Jun

   2025

   Bib

   @article{paper2,
     title = {Second-Order Bounds for [0,1]-Valued Regression via Betting Loss},
     author = {Li, Yinan and Huang, Ethan and Yoon, Sungjoon and Jun, Kwang-Sung},
     year = {2025}
   }

3. When Can Proxies Improve the Sample Complexity of Preference Learning?

   Yuchen Zhu, Daniel Souza, Zhengyan Shi, Mengyue Yang, Pasquale Minervini, Matt J. Kusner, and Alexander Nicholas D’Amour

   2025

   Bib

   @article{paper4,
     title = {When Can Proxies Improve the Sample Complexity of Preference Learning?},
     author = {Zhu, Yuchen and Augusto de Souza, Daniel and Shi, Zhengyan and Yang, Mengyue and Minervini, Pasquale and Kusner, Matt J. and D'Amour, Alexander Nicholas},
     year = {2025}
   }

4. The Importance of Online Data: Understanding Preference Fine-tuning via Coverage

   Yuda Song, Gokul Swamy, Aarti Singh, Drew Bagnell, and Wen Sun

   2025

   Bib

   @article{paper6,
     title = {The Importance of Online Data: Understanding Preference Fine-tuning via Coverage},
     author = {Song, Yuda and Swamy, Gokul and Singh, Aarti and Bagnell, Drew and Sun, Wen},
     year = {2025}
   }

5. ROC-Climbing: Test-time scaling with imperfect verifiers

   Florian E. Dorner, Yatong Chen, Andre F Cruz, and Fanny Yang

   2025

   Bib

   @article{paper8,
     title = {ROC-Climbing: Test-time scaling with imperfect verifiers},
     author = {Dorner, Florian E. and Chen, Yatong and Cruz, Andre F and Yang, Fanny},
     year = {2025}
   }

6. All Roads Lead to Likelihood: The Value of Reinforcement Learning in Fine-Tuning

   Gokul Swamy, Sanjiban Choudhury, Wen Sun, Steven Wu, and Drew Bagnell

   2025

   Bib

   @article{paper9,
     title = {All Roads Lead to Likelihood: The Value of Reinforcement Learning in Fine-Tuning},
     author = {Swamy, Gokul and Choudhury, Sanjiban and Sun, Wen and Wu, Steven and Bagnell, Drew},
     year = {2025}
   }

7. Parameter Efficient Model Merging

   Margalit Glasgow, Alexander Rakhlin, Sasha Voitovych, and Fan Chen

   2025

   Bib

   @article{paper10,
     title = {Parameter Efficient Model Merging},
     author = {Glasgow, Margalit and Rakhlin, Alexander and Voitovych, Sasha and Chen, Fan},
     year = {2025}
   }

8. Quantitative Bounds for Length Generalization in Transformers

   Zachary Izzo, Eshaan Nichani, and Jason D. Lee

   2025

   Bib

   @article{paper11,
     title = {Quantitative Bounds for Length Generalization in Transformers},
     author = {Izzo, Zachary and Nichani, Eshaan and Lee, Jason D.},
     year = {2025}
   }

9. Accelerating Nash Learning from Human Feedback via Mirror Prox

   Daniil Tiapkin, Daniele Calandriello, Denis Belomestny, Eric Moulines, Alexey Naumov, Kashif Rasul, Michal Valko, and Pierre Menard

   2025

   Bib

   @article{paper12,
     title = {Accelerating Nash Learning from Human Feedback via Mirror Prox},
     author = {Tiapkin, Daniil and Calandriello, Daniele and Belomestny, Denis and Moulines, Eric and Naumov, Alexey and Rasul, Kashif and Valko, Michal and Menard, Pierre},
     year = {2025}
   }

10. Low-rank fine-tuning lies between lazy training and feature learning

    Arif Kerem Dayi and Sitan Chen

    2025

    Bib

    @article{paper13,
      title = {Low-rank fine-tuning lies between lazy training and feature learning},
      author = {Dayi, Arif Kerem and Chen, Sitan},
      year = {2025}
    }

11. SharedRep-RLHF: A Shared Representation Approach to RLHF with Diverse Preferences

    Arpan Mukherjee, Marcello Bullo, and Deniz Gunduz

    2025

    Bib

    @article{paper14,
      title = {SharedRep-RLHF: A Shared Representation Approach to RLHF with Diverse Preferences},
      author = {Mukherjee, Arpan and Bullo, Marcello and Gunduz, Deniz},
      year = {2025}
    }

12. Certifiably Safe Post-Training

    Pierre Fasterling, Leo Elmecker-Plakolm, Philip Sosnin, Calvin Tsay, and Matthew Robert Wicker

    2025

    Bib

    @article{paper15,
      title = {Certifiably Safe Post-Training},
      author = {Fasterling, Pierre and Elmecker-Plakolm, Leo and Sosnin, Philip and Tsay, Calvin and Wicker, Matthew Robert},
      year = {2025}
    }

13. Weak-to-Strong Generalization Even in Random Feature Networks, Provably

    Marko Medvedev, Kaifeng Lyu, Dingli Yu, Sanjeev Arora, Zhiyuan Li, and Nathan Srebro

    2025

    Bib

    @article{paper16,
      title = {Weak-to-Strong Generalization Even in Random Feature Networks, Provably},
      author = {Medvedev, Marko and Lyu, Kaifeng and Yu, Dingli and Arora, Sanjeev and Li, Zhiyuan and Srebro, Nathan},
      year = {2025}
    }

14. PILAF: Optimal Human Preference Sampling for Reward Modeling

    Yunzhen Feng, Ariel Kwiatkowski, Kunhao Zheng, Yaqi Duan, and Julia Kempe

    2025

    Bib

    @article{paper17,
      title = {PILAF: Optimal Human Preference Sampling for Reward Modeling},
      author = {Feng, Yunzhen and Kwiatkowski, Ariel and Zheng, Kunhao and Duan, Yaqi and Kempe, Julia},
      year = {2025}
    }

15. Imitation Learning and Supervised Fine Tuning with Deterministic vs Stochastic Policies and Generators

    Nirmit Joshi, Gene Li, Gal Vardi, and Nathan Srebro

    2025

    Bib

    @article{paper18,
      title = {Imitation Learning and Supervised Fine Tuning with Deterministic vs Stochastic Policies and Generators},
      author = {Joshi, Nirmit and Li, Gene and Vardi, Gal and Srebro, Nathan},
      year = {2025}
    }