ELLIS Institute Finland at ICLR and AISTATS 2026
ELLIS Institute Finland's researchers are sharing their results at two major conferences in the AI and ML fields, in April and May 2026.
ICLR, the International Conference on Learning Representations, will be held in Rio de Janeiro, Brazil, April 23-27, 2026.
AISTATS, the Conference on Artificial Intelligence and Statistics, will be in Tangier, Morocco, May 2-5, 2026.
Oral contributions at ICLR
| Rafał Karczewski, Markus Heinonen, Alison Pouplin, Søren Hauberg, Vikas Garg | The Spacetime of Diffusion Models: An Information Geometry Perspective |
Posters at ICLR
| Petrus Mikkola, Luigi Acerbi, Arto Klami | Score-Based Density Estimation from Pairwise Comparisons |
| Aki Rehn, Linzh Zhao, Mikko A. Heikkilä, Antti Honkela | On Optimal Hyperparameters for Differentially Private Deep Transfer Learning |
| Hengyu Luo, Raúl Vázquez, Timothee Mickus, Filip Ginter, Jörg Tiedemann | Knowledge Distillation as Decontamination? Revisiting the “Data Laundering” Concern |
| Yang Yang, Severi Rissanen, Paul E. Chang, Nasrulloh Loka, Daolang Huang, Arno Solin, Markus Heinonen, Luigi Acerbi | PriorGuide: Test-Time Prior Adaptation for Simulation-Based Inference |
| Conor Hassan, Nasrulloh Loka, Cen-You Li, Daolang Huang, Paul E. Chang, Yang Yang, Francesco Silvestrin, Samuel Kaski, Luigi Acerbi | Efficient Autoregressive Inference for Transformer Probabilistic Models |
| Gauri Pradhan, Joonas Jälkö, Santiago Zanella-Béguelin, Antti Honkela | Beyond Membership: Limitations of Add/Remove Adjacency in Differential Privacy |
| Bohao Xing, Xin Liu, Guoying Zhao, Chengyu Liu, Xiaolan Fu, Heikki Kälviäinen | EmotionHallucer: Evaluating Emotion Hallucinations in Multimodal Large Language Models |
| Zheng Lian, Licai Sun, Lan Chen, Haoyu Chen, Zebang Cheng, Fan Zhang, Ziyu Jia, Ziyang Ma, Fei Ma, Xiaojiang Peng, Jianhua Tao | EmoPrefer: Can Large Language Models Understand Human Emotion Preferences? |
| Anton Baumann, Rui Li, Marcus Klasson, Santeri Mentu, Shyamgopal Karthik, Zeynep Akata, Arno Solin, Martin Trapp | Post-hoc Probabilistic Vision-Language Models |
| Andi Zhang, Xuan Ding, Steven McDonagh, Samuel Kaski | Concept-based Adversarial Attack: a Probabilistic Perspective |
| Peter Blohm, Vikas Garg | Hot Fuzz: Temperature-Tunable Composition of Diffusion models with Fuzzy Logic |
| Amirreza Akbar, Amari H. Souza, Vikas Garg | The logical expressiveness of topological neural networks |
| Fernando Ruiz-Mazo, Vikas Garg | Frozen Priors, Fluid Forecasts: Prequential Uncertainty for Low-Data Deployment with Pretrained Generative Models |
| Mattie Ji, Indradyumna Roy, Vikas Garg | Hourglass Persistence for Graphs, Simplices, and Cells |
| Mohammad Hassan Vali, Tom Bäckström, Arno Solin | DiVeQ: Differentiable Vector Quantization Using the Reparameterization Trick |
| Yi Zhao, Aidan Scannell, Wenshuai Zhao, Yuxin Hou, Tianyu Cui, Le Chen, Dieter Büchler, Arno Solin, Juho Kannala, Joni Pajarinen | Efficient Reinforcement Learning by Guiding World Models with Non-Curated Data |
| Xinyu Zhang, Conor Hassan, Julien Martinelli, Daolang Huang, Samuel Kaski | Task-Agnostic Amortized Multi-Objective Optimization |