28.11.2025
47 Accepted Papers (38 Main, and 9 Workshops)
39th Conference on Neural Information Processing Systems, San Diego, CA, USA, Nov 30-Dec 07, 2025
We are happy to announce that MCML researchers have contributed a total of 47 papers to NeurIPS 2025: 38 Main, and 9 Workshop papers. Congrats to our researchers!
Main Track (38 papers)
Improving the Generation and Evaluation of Synthetic Data for Downstream Medical Causal Inference.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
FlowFeat: Pixel-Dense Embedding of Motion Profiles.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. Spotlight Presentation. To be published. Preprint available. URL
NOVA: A Benchmark for Rare Anomaly Localization and Clinical Reasoning in Brain MRI.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
Explaining Similarity in Vision-Language Encoders with Weighted Banzhaf Interactions.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
Revisiting Glorot Initialization for Long-Range Linear Recurrences.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
Non-Singularity of the Gradient Descent Map for Neural Networks with Piecewise Analytic Activations.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. Spotlight Presentation. To be published. Preprint available. URL
ConceptScope: Characterizing Dataset Bias via Disentangled Visual Concepts.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
Adjusted Count Quantification Learning on Graphs.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
Anomaly Detection by an Ensemble of Random Pairs of Hyperspheres.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
OrthoLoC: UAV 6-DoF Localization and Calibration Using Orthographic Geodata.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. Oral Presentation. To be published. Preprint available. URL GitHub
Improving Perturbation-based Explanations by Understanding the Role of Uncertainty Calibration.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. Spotlight Presentation. To be published. Preprint available. URL
Noise Hypernetworks: Amortizing Test-Time Compute in Diffusion Models.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL GitHub
Treatment Effect Estimation for Optimal Decision-Making.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
Orthogonal Survival Learners for Estimating Heterogeneous Treatment Effects from Time-to-Event Data.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
Video Perception Models for 3D Scene Synthesis.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
Differentiable Sparsity via D-Gating: Simple and Versatile Structured Penalization.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. Spotlight Presentation. To be published. Preprint available. URL
Performative Validity of Recourse Explanations.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
TreeGen: A Bayesian Generative Model for Hierarchies.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
ProSpero: Active Learning for Robust Protein Design Beyond Wild-Type Neighborhoods.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
ResponseRank: Data-Efficient Reward Modeling through Preference Strength Learning.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. URL
Neural Emulator Superiority: When Machine Learning for PDEs Surpasses its Training Data.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. URL GitHub
Credal Prediction based on Relative Likelihood.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. Spotlight Presentation. To be published. Preprint available. URL
Kernel conditional tests from learning-theoretic bounds.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
LLM-Driven Treatment Effect Estimation Under Inference Time Text Confounding.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. arXiv
Measuring Fingerprints of Web-filtered Text Datasets and Fingerprint Propagation Through Training.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. Spotlight Presentation. To be published. Preprint available. URL
Personalized Exercise Recommendation with Semantically-Grounded Knowledge Tracing.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
EgoExOR: An Ego-Exo-Centric Operating Room Dataset for Surgical Activity Understanding.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
Sparse Autoencoders Learn Monosemantic Features in Vision-Language Models.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
DisMo: Disentangled Motion Representations for Open-World Motion Transfer.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. Spotlight Presentation. To be published. Preprint available. URL
Failure Prediction at Runtime for Generative Robot Policies.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. arXiv GitHub
Cross-fluctuation phase transitions reveal sampling dynamics in diffusion models.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
Conformal Prediction for Causal Effects of Continuous Treatments.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
DCAD-2000: A Multilingual Dataset across 2000+ Languages with Data Cleaning as Anomaly Detection.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. URL
Modeling Microenvironment Trajectories on Spatial Transcriptomics with NicheFlow.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
Equivariance by Contrast: Identifiable Equivariant Embeddings from Unlabeled Finite Group Actions.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
Learning Generalizable Shape Completion with SIM(3) Equivariance.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
Refusal Direction is Universal Across Safety-Aligned Languages.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
Learning Linear Attention in Polynomial Time.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. Oral Presentation. To be published. Preprint available. URL
Workshops (9 papers)
Deep Research Brings Deeper Harm.
ReliableML @NeurIPS 2025 - Workshop on Reliable ML from Unreliable Data at the 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. arXiv GitHub
Does Machine Unlearning Truly Remove Knowledge?
Lock-LLM @NeurIPS 2025 - Lock-LLM Workshop: Prevent Unauthorized Knowledge Use from Large Language Models at the 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
Human Uncertainty-Aware Reliable Data Selection and Efficient Annotation for Visual Question Answering.
ReliableML @NeurIPS 2025 - Workshop on Reliable ML from Unreliable Data at the 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
Reasoning with Fewer Eyes: Efficient Visual Token Withdrawal for Multimodal Reasoning.
ER @NeurIPS 2025 - Workshop on Efficient Reasoning at the 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
Predicting symbolic ODEs from multiple trajectories.
ML4PS @NeurIPS 2025 - Workshop on Machine Learning and the Physical Sciences at the 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. PDF
Hierarchical Graph Networks for Forecasting Terrestrial Water Storage Anomalies.
ML4PS @NeurIPS 2025 - Workshop on Machine Learning and the Physical Sciences at the 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. PDF
SLAyiNG: Towards Queer Language Processing.
QueerInAI @NeurIPS 2025 - Queer in AI Workshop at the 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. arXiv
How reliable are treatment effects in clinical trials with dropout?
CauScien @NeurIPS 2025 - Workshop on Uncovering Causality in Science at the 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
GroundedPRM: Tree-Guided and Fidelity-Aware Process Reward Modeling for Step-Level Reasoning.
SEA @NeurIPS 2025 - Workshop on Scaling Environments for Agents at the 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. To be published. Preprint available. URL
Related
©BAdW / Karl Neunert
08.12.2025
Tom Sterkenburg Wins Karl-Heinz Hoffmann Prize of the Bavarian Academy of Sciences
MCML JRG Leader Tom Sterkenburg receives the Karl-Heinz Hoffmann Prize of the BAdW for his interdisciplinary research.
©Juli Eberle / TUM / ediundsepp Gestaltungsgesellschaft
04.12.2025
World’s First Complete 3D Model of All Buildings Released
Xiaoxiang Zhu’s team releases GlobalBuildingAtlas, a high-res 3D map of 2.75B buildings for advanced urban and climate analysis.
04.12.2025
When to Say "I’m Not Sure": Making Language Models More Self-Aware
ICLR 2025 research by the groups of David Rügamer, and Bernd Bischl introduces methods to make LLMs more reliable by expressing uncertainty.
©MCML
01.12.2025
Research Stay at Princeton University
Abdurahman Maarouf spent three months at Princeton with the AI X-Change Program, advancing causal ML and studying short-form video platform effects.