Home | News

11.07.2025

Tiny logo
Teaser image to MCML at ICML 2025

MCML at ICML 2025

27 Accepted Papers (21 Main, and 6 Workshops)

42nd International Conference on Machine Learning, Vancouver, Canada, Jul 13-19, 2025

We are happy to announce that MCML researchers have contributed a total of 27 papers to ICML 2025: 21 Main, and 6 Workshop papers. Congrats to our researchers!

Main Track (21 papers)

W. Durani • T. Nitzl • C. Plant • C. Böhm
Weakly Supervised Anomaly Detection via Dual-Tailed Kernel.
ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL

X. Feng • Z. Jiang • T. Kaufmann • E. Hüllermeier • P. Weng • Y. Zhu
Comparing Comparisons: Informative and Easy Human Feedback with Distinguishability Queries.
ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL

U. Fischer Abaigar • C. Kern • J. Perdomo
The Value of Prediction in Identifying the Worst-Off.
ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. Spotlight Presentation. Outstanding Paper Award. URL

P. Fatemi • E. Sharifian • M. H. Yassaee
A New Approach to Backtracking Counterfactual Explanations: A Unified Causal Framework for Efficient Model Interpretability.
ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL

L. Götz • M. Kollovieh • S. Günnemann • L. Schwinn
Efficient Time Series Processing for Transformers and State-Space Models through Token Merging.
ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL

S. Karnik • A. Veselovska • M. Iwen • F. Krahmer
Implicit Regularization for Tubal Tensor Factorizations via Gradient Descent.
ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL

W. Lai • A. Fraser • I. Titov
Joint Localization and Activation Editing for Low-Resource Fine-Tuning.
ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL

M. Lienen • A. Saydemir • S. Günnemann
UnHiPPO: Uncertainty-aware Initialization for State Space Models.
ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL

A. Modarressi • H. Deilamsalehy • F. Dernoncourt • T. Bui • R. A. Rossi • S. Yoon • H. Schütze
NoLiMa: Long-Context Evaluation Beyond Literal Matching.
ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL URL

S. Müller • A. Reuter • N. Hollmann • D. Rügamer • F. Hutter
Position: The Future of Bayesian Prediction Is Prior-Fitted.
ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL

D. A. Nguyen • E. Araya • A. Fono • G. Kutyniok
Time to Spike? Understanding the Representational Power of Spiking Neural Networks in Discrete Time.
ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL

T. Pielok • B. Bischl • D. Rügamer
Revisiting Unbiased Implicit Variational Inference.
ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL

A. Reuter • T. G. J. Rudner • V. Fortuin • D. Rügamer
Can Transformers Learn Full Bayesian Inference in Context?
ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL

J. Schweisthal • D. Frauen • M. Schröder • K. Heß • N. Kilbertus • S. Feuerriegel
Learning Representations of Instruments for Partial Identification of Treatment Effects.
ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL

N. Schneider • L. Lorch • N. Kilbertus • B. Schölkopf • A. Krause
Generative Intervention Models for Causal Perturbation Modeling.
ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL

R. Schulte • D. Rügamer • T. Nagler
Adjustment for Confounding using Pre-Trained Representations.
ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL

A. Soleymani • B. Tahmasebi • S. Jegelka • P. Jaillet
Learning with Exact Invariances in Polynomial Time.
ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL

D. Tramontano • Y. Kivva • S. Salehkaleybar • N. Kiyavash • M. Drton
Causal Effect Identification in lvLiNGAM from Higher-Order Cumulants.
ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL

L. Thede • K. Roth • M. Bethge • Z. Akata • T. Hartvigsen
WikiBigEdit: Understanding the Limits of Lifelong Knowledge Editing in LLMs.
ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL

A. Uselis • A. Dittadi • S. J. Oh
Does Data Scaling Lead to Visual Compositional Generalization?
ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL GitHub

J. Zausinger • L. Pennig • A. Kozina • S. Sdahl • J. Sikora • A. Dendorfer • T. Kuznetsov • M. Hagog • N. Wiedemann • K. Chlodny • V. Limbach • A. Ketteler • T. Prein • V. M. Singh • M. M. Danziger • J. Born
Regress, Don't Guess -- A Regression-like Loss on Number Tokens for Language Models.
ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL GitHub

Workshops (6 papers)

J. von Berg • A. Fono • M. Datres • S. Maskey • G. Kutyniok
The Price of Robustness: Stable Classifiers Need Overparameterization.
HiLD @ICML 2025 - Workshop on High-dimensional Learning Dynamics at the 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL

F. Kreuter
Adaptive Alignment: Designing AI for a Changing World - Frauke Kreuter.
Invited Talk @ICML 2025 - 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. Invited Talk. URL

F. Kiwitt • B. Tahmasebi • S. Jegelka
Symmetries in Weight Space Learning: To Retain or Remove?
HiLD @ICML 2025 - Workshop on High-dimensional Learning Dynamics at the 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL

Z. Li • X. Han • Y. Li • N. Strauß • M. Schubert
DAWM: Diffusion Action World Models for Offline Reinforcement Learning via Action-Inferred Transitions.
WM @ICML 2025 - Workshop on Building Physically Plausible World Models at the 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. arXiv

P. Spohn • L. Girrbach • J. Bader • Z. Akata
Align-then-Unlearn: Embedding Alignment for LLM Unlearning.
MUGen @ICML 2025 - Workshop on Machine Unlearning for Generative AI at the 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. URL

L. Xu • M. Sarkar • A. I. Lonappan • Í. Zubeldia • P. Villanueva-Domingo • S. Casas • C. Fidler • C. Amancharla • U. Tiwari • A. Bayer • C. A. Ekioui • M. Cranmer • A. Dimitrov • J. Fergusson • K. Gandhi • S. Krippendorf • A. Laverick • J. Lesgourgues • A. Lewis • T. Meier • B. Sherwin • K. Surrao • F. Villaescusa-Navarro • C. Wang • X. Xu • B. Bolliet
Open Source Planning & Control System with Language Agents for Autonomous Scientific Discovery.
ML4Astro @ICML 2025 - Machine Learning for Astrophysics at the 42nd International Conference on Machine Learning. Vancouver, Canada, Jul 13-19, 2025. PDF

#research #top-tier-work #akata #bauer-s #bischl #drton #feuerriegel #fornasier #fortuin #fraser #guennemann #huellermeier #jegelka #kern #kilbertus #krahmer #kreuter #kutyniok #nagler #rueckert #ruegamer #schubert #schuetze #seidl #sra

Related

Link to Digdeep Podcast: Career leaps and comfort zones – what does it take for a successful career, Daniela Gerd tom Markotten?

13.11.2026

Digdeep Podcast: Career Leaps and Comfort Zones – What Does It Take for a Successful Career, Daniela Gerd Tom Markotten?

In the new episode of #digdeep, Daniela Gerd tom Markotten talks about technology, speed and career advancements.

Read more
Link to One Toolbox for Many Kinds of Uncertainty

12.11.2026

One Toolbox for Many Kinds of Uncertainty

MCML researchers introduce probly, a unified Python package for uncertainty-aware machine learning, presented orally at NeurIPS 2026.

Read more
Link to Enkelejda Kasneci: The Real Value of AI in Education Is More Time for Humans

10.11.2026

Enkelejda Kasneci: The Real Value of AI in Education Is More Time for Humans

Enkelejda Kasneci explores how AI can provide personalized learning support while keeping human interaction and student agency at the center.

Read more
Link to Outstanding Paper Award at EMNLP 2026 for Work on Accent Bias in Audio Language Models

09.11.2026

Outstanding Paper Award at EMNLP 2026 for Work on Accent Bias in Audio Language Models

A paper co-authored by MCML PI Valentin Hofmann on accent bias in audio language models won an Outstanding Paper Award at EMNLP 2026.

Read more
Link to Best Thesis Award for Sohir Maskey

06.11.2026

Best Thesis Award for Sohir Maskey

Former MCML Junior Member Sohir Maskey received the Best Thesis Award of the DFG Priority Programme 2298 for his PhD on graph neural networks.

Read more
Back to Top