LIONS @ ICLR 2025

© 2025 EPFL

© 2025 EPFL

7 papers from the LIONS group have been accepted to the 2025 edition of ICLR conference. Congratulations to all!
This year ICLR conference will be held in Singapore, from 24-28 April.

Below is a list of ICLR 2025 papers with at least one LIONS author:

  1. Adversarial Training for Defense Against Label Poisoning Attacks by Melis Ilayda Bal, Volkan Cevher, Michael Muehlebach
  2. Certified Robustness Under Bounded Levenshtein Distance by Elias Abad Rocamora, Grigorios Chrysos, Volkan Cevher
  3. Faster Inference of Flow-Based Generative Models via Improved Data-Noise Coupling by Aram Davtyan, Leello Tadesse Dadi, Volkan Cevher, Paolo Favaro
  4. Efficient Interpolation between Extragradient and Proximal Methods for Weak MVIs by Thomas Pethick, Ioannis Mavrothalassitis, Volkan Cevher
  5. Addressing Label Shift in Distributed Learning via Entropy Regularization​ by Zhiyuan Wu, Changkyu Choi, Xiangcheng Cao, Volkan Cevher, Ali Ramezani-Kebrya
  6. How do students become teachers: A dynamical analysis for two-layer neural networks by Zhenyu Zhu, Fanghui Liu, Volkan Cevher
  7. Quantum-PEFT: Ultra parameter-efficient fine-tuning by Toshiaki Koike-Akino, Francesco Tonin, Yongtao Wu, Zhengqing Wu, Leyla Naz Candogan, Volkan Cevher