Accepted Papers at ICLR and AISTATS 2026

Eight papers from the Helsinki Probabilistic Machine Learning Lab have been accepted at major conferences this January.

Five papers will be presented at the International Conference on Learning Representations (ICLR 2026) and three papers at the International Conference on Artificial Intelligence and Statistics (AISTATS 2026):
 

  1. Score-Based Density Estimation from Pairwise Comparisons by Petrus Mikkola, Luigi Acerbi, Arto Klami,
    International Conference on Learning Representations (ICLR 2026). []
     
  2. PriorGuide: Test-Time Prior Adaptation for Simulation-Based Inference by Yang Yang, Severi Rissanen, Paul E. Chang, Nasrulloh Loka, Daolang Huang, Arno Solin, Markus Heinonen, Luigi Acerbi,
    International Conference on Learning Representations (ICLR 2026). []
     
  3. Efficient Autoregressive Inference for Transformer Probabilistic Models by Conor Hassan*, Nasrulloh Loka*, Cen-You Li, Daolang Huang, Paul E. Chang, Yang Yang, Francesco Silvestrin, Samuel Kaski, Luigi Acerbi,
    International Conference on Learning Representations (ICLR 2026). []
     
  4. On Optimal Hyperparameters for Differentially Private Deep Transfer Learning by Aki Rehn, Linzh Zhao, Mikko A. Heikkilä, Antti Honkela,
    International Conference on Learning Representations (ICLR 2026). []
     
  5. Beyond Membership: Limitations of Add/Remove Adjacency in Differential Privacy by Gauri Pradhan, Joonas Jälkö, Santiago Zanella-Bèguelin, Antti Honkela,
    International Conference on Learning Representations (ICLR 2026). []
     
  6. Simplex-to-Euclidean Bijections for Categorical Flow Matching by Bernardo Williams, Victor M Yeom-Song, Marcelo Hartmann, Arto Klami,
    International Conference on Artificial Intelligence and Statistics (AISTATS 2026). []
     
  7. Learning geometry and topology via multi-chart flows by Hanlin Yu, Søren Hauberg, Marcelo Hartmann, Arto Klami, Georgios Arvanitidis,
    International Conference on Artificial Intelligence and Statistics (AISTATS 2026). []
     
  8. On the Identifiability of Tensor Ranks via Prior Predictive Matching by Eliezer de Souza da Silva, Arto Klami, Diego Mesquita, Iñigo Urteaga,
    International Conference on Artificial Intelligence and Statistics (AISTATS 2026). []
     


See our page for more information.