LIONS @ NeurIPS 2026
NeurIPS logo© 2026 EPFL
11 papers from the LIONS group have been accepted to the 2026 edition of NeurIPS conference. Congratulations to all!
This year, the 40th Annual Conference on Neural Information Processing Systems conference will be held in Sydney, Australia from 6 to 12 December.
- Optimistic Dual Averaging Unifies Modern Optimizers by Thomas Pethick, Wanyun Xie, Roman Machacek, Volkan Cevher
- AcceleGrad#: Adaptive Geometry-Aware Acceleration by Hanka Goralija*, Francesco Tonin*, Kimon Antonakopoulos, Alp Yurtsever, Volkan Cevher
- Constrained Stochastic Spectral Preconditioning Converges for Nonconvex Objectives by Konstantinos Oikonomidis, Jan Quan, Kimon Antonakopoulos, Tony Silveti-Falls, Volkan Cevher*, Panagiotis Patrinos*
- When Does Interaction Help? Representational Tradeoffs in Value-Based Imitation Learning by Luca Viano, Antoine Moulin, Audrey Huang, Volkan Cevher, Philip Amortila, Dylan J Foster
- Why Routers Freeze: Infinite Width Learning Dynamics for Mixture of Experts by Anish Dhir, Volkan Cevher, Leena Chennuru Vankadara
- Split the Differences, Pool the Rest: Provably Efficient Multi-Objective Imitation by Ziyad Sheebaelhamd, Luca Viano, Volkan Cevher, Claire Vernade
- Raven: High-Recall Sequence Modeling via Sparse Memory Routing by Arshia Afzal, Aviv Bick, Eric P. Xing, Volkan Cevher, Albert Gu
- Attention-Discounted Adaptive Sampler for Masked Diffusion Language Models by Yusuf Sahin, Ahmed R. Saikia, Volkan Cevher, Paolo Favaro
- Rescaled Asynchronous SGD: Optimal Distributed Optimization under Data and System Heterogeneity by Ammar Mahran, Artavazd Maranjyan, Peter Richtárik
- LOSCAR-SGD: Local SGD with Communication-Computation Overlap and Delay-Corrected Sparse Model Averaging by Yassine Maziane, Ammar Mahran, Artavazd Maranjyan, Peter Richtárik
- TempoPFN: Synthetic Pre-training of Linear RNNs for Zero-shot Time Series Forecasting by Vladyslav Moroshan, Julien Siems, Arber Zela, Timur Carstensen, Frank Hutter