ICML 2026 accepted papers on Probabilistic and scientific modeling
613 of the 6,341 papers accepted at ICML 2026 (9.7%) are on Probabilistic and scientific modeling, compared with 10.2% at ICML 2025. They fall into 8 topics.
See ICML 2026 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Molecules and proteins · 214 papers
- SLAE: Strictly Local All-atom Environment for Protein Representation (2 citations)
- DISSOLVR: An Interpretable and Fast Framework for Aqueous and Organic Solubility Prediction (1 citation)
- PluRel: Synthetic Data unlocks Scaling Laws for Relational Foundation Models (1 citation)
- Protein Language Model Embeddings Improve Generalization of Implicit Transfer Operators (1 citation)
- A Geometric Lens on Physics-Aligned Data Compression
- A Kinetic Energy Perspective of Flow Matching
- A Machine-Learned Comorbidity Index
- A Unified Density Operator View of Flow Control and Merging
- Accurate Evaluation of Quickest Changepoint Detectors via Non-parametric Survival Analysis
- Adaptive Protein Tokenization
- Adversarially Robust Control of Conditional Value-at-Risk via Rockafellar-Uryasev Conformal Inference
- Agentic Framework for Epidemiological Modeling
and 202 more in the interactive map.
Variational and generative sampling · 96 papers
- $f$-Trajectory Balance: A Loss Family for Tuning GFlowNets, Generative Models, and LLMs with Off- and On-Policy Data
- A Diffusive Classification Loss for Learning Energy-based Generative Models
- A hitchhiker's guide to Poisson gradient estimation
- A Probabilistic Framework for LLM-Based Model Discovery
- A Time-Reparameterized Cumulative Intensity Extrapolation Sampler for Discrete Flow Matching
- A Unifying View of Variational Generative Wasserstein Flows
- Adaptive Estimation and Inference in Semi-parametric Heterogeneous Clustered Multitask Learning via Neyman Orthogonality
- Amortized Simulation-Based Inference in Generalized Bayes via Neural Posterior Estimation
- Analytic Bijections for Smooth and Interpretable Normalizing Flows
- Autoregressive Boltzmann Generators
- Beyond Softmax: A Natural Parameterization for Categorical Random Variables
- Can Microcanonical Langevin Dynamics Leverage Mini-Batch Gradient Noise?
and 84 more in the interactive map.
Dynamical systems · 77 papers
- Constrained hybrid modelling to predict microbial dynamics and organic matter turnover in soil systems (1 citation)
- Meta-learning Structure-Preserving Dynamics (1 citation)
- A Call to Lagrangian Action: Learning Population Mechanics from Temporal Snapshots
- A Factorized Low-Rank RNN Framework for Uncovering Independent Neural Latent Dynamics and Connectivity
- Adaptive Symmetry Discovery for Dynamical System Identification
- Beyond Continuity: Simulation-free Reconstruction of Discrete Branching Dynamics from Single-cell Snapshots
- Cheap2Rich: A Multi-Fidelity Framework for Data Assimilation and System Identification of Multiscale Physics - Rotating Detonation Engines
- Continuous Variable Hamiltonian Learning at Heisenberg Limit via Displacement-Random Unitary Transformation
- DAISI: Data Assimilation with Inverse Sampling using Stochastic Interpolants
- DRL-STAF: A Deep Reinforcement Learning Framework for State-Aware Forecasting of Complex Multivariate Hidden Markov Processes
- Dynamic Compression Flows for Neuroscience Data
- Dynamics of neural scaling laws in random feature regression with powerlaw-distributed kernel eigenvalues
and 65 more in the interactive map.
Uncertainty and Gaussian processes · 75 papers
- Cooperative Variance Estimation and Bayesian Neural Networks for Disentangling Aleatoric and Epistemic Uncertainties (2 citations)
- Spatial Conformal Inference through Localized Quantile Regression (1 citation)
- Accuracy and Normalized Accuracy under Length Bias: Analysis, Guidelines, and a Bayesian Alternative
- Active Timepoint Selection for Learning Measure-Valued Trajectories
- Adaptive Querying with AI Persona Priors
- ANCHOR: Abductive Network Construction with Hierarchical Orchestration for Reliable Probability Inference in Large Language Models
- BALLAST: Bayesian Active Learning with Look-ahead Amendment for Sea-drifter Trajectories under Spatio-Temporal Vector Fields
- Bayesian model selection and misspecification testing in imaging inverse problems only from noisy and partial measurements
- Bayesian Rain Field Reconstruction using Commercial Microwave Links and Diffusion Model Priors
- Bayesian-LoRA: Probabilistic Low-Rank Adaptation of Large Language Models
- Belief Propagation Converges to Gaussian Distributions in Sparsely-Connected Factor Graphs
- BLIPs: Bayesian Learned Interatomic Potentials
and 63 more in the interactive map.
Causal discovery · 54 papers
- A Recursive Decomposition Framework for Causal Structure Learning in the Presence of Latent Variables
- Causal Fine-Tuning under Latent Confounded Shift
- Causal Identification from Counterfactual Data: Completeness and Bounding Results
- Causal Modeling of Selection in Evolution
- Causal Preference Elicitation
- CausalProfiler: Generating Synthetic Benchmarks for Rigorous and Transparent Evaluation of Causal Machine Learning
- DISCO: Mitigating Bias in Deep Learning with Conditional Distance Correlation
- Distributionally Robust Causal Abstractions
- Dynamics Within Latent Chain-of-Thought: An Empirical Study of Causal Structure
- Evaluating Bivariate Causal Statements Based on Mutual Compatibility
- Exactly Computing do-Shapley Values
- Formalizing and Falsifying Causal Pathways of Rare Events
and 42 more in the interactive map.
Differential equations and PDEs · 49 papers
- Neuro-Symbolic AI for Analytical Solutions of Differential Equations (1 citation)
- A Dirac-Frenkel-Onsager Principle: Instantaneous Residual Minimization with Gauge Momentum for Nonlinear Parametrizations of PDE Solutions
- ANTiC: Adaptive Neural Temporal In Situ Compressor
- Di-BiLPS: Denoising induced Bidirectional Latent-PDE-Solver under Sparse Observations
- Discontinuous Galerkin Neural Operator for Pathology Defocus Deblurring
- Discovering Scaling Exponents with Physics-Informed Müntz-Szász Networks
- Efficient Neural Controlled Differential Equations via Attentive Kernel Smoothing
- EqGINO: Equivariant Geometry-Informed Fourier Neural Operators for 3D PDEs
- Expanding the Chaos: Neural Operator for Stochastic (Partial) Differential Equations
- FluxNet: Learning Capacity-Constrained Local Transport Operators for Conservative and Bounded PDE Surrogates
- Foundation Inference Models for Ordinary Differential Equations
- From Basis to Basis: Gaussian Particle Representation for Interpretable PDE Operators
and 37 more in the interactive map.
Treatment effects · 29 papers
- Addressing Instrument-Outcome Confounding in Mendelian Randomization through Representation Learning
- Budgeted Active Experimentation for Treatment Effect Estimation from Observational and Randomized Data
- Causal Matrix Completion under Multiple Treatments via Mixed Synthetic Nearest Neighbors
- Conditional Distributional Treatment Effects: Doubly Robust Estimation and Testing
- DeepBlip: Estimating Conditional Average Treatment Effects Over Time
- Design-Based Anytime-Valid Inference for Randomized Experiments with Delayed Outcomes and Staggered Entry
- Estimating Continuous Treatment Effects with Two-Stage Kernel Ridge Regression
- Estimation of Treatment Effects Under Nonstationarity via the Truncated Policy Gradient Estimator
- Feasible Fusion: Constrained Joint Estimation under Structural Non-Overlap
- Frequentist Consistency of Prior-Data Fitted Networks for Causal Inference
- Learning Treatment Representations for Downstream Instrumental Variable Regression
- Local Covariate Selection for Average Causal Effect Estimation without Pretreatment and Causal Sufficiency Assumptions
and 17 more in the interactive map.
Bayesian optimization · 19 papers
- $f$-Divergence Regularized RLHF: Two Tales of Sampling and Unified Analyses
- ALAS: Additive Learnable Alpha-Stable Kernels for Flexible Bayesian Optimization
- Bridging Spherical Black-Box Optimizers
- Cost-aware Stopping for Bayesian Optimization
- Divide and Learn: Multi-Objective Combinatorial Optimization at Scale
- Few-Shot Design Optimization by Exploiting Auxiliary Information
- Goal-Oriented Lower-Tail Calibration of Gaussian Processes for Bayesian Optimization
- Local Constrained Bayesian Optimization
- Many Needles in a Haystack: Active Hit Discovery for Perturbation Experiments
- Regime-Adaptive Bayesian Optimization via Dirichlet Process Mixtures of Gaussian Processes
- Robust Bayesian Optimisation with Unbounded Corruptions
- Support-Proximity Augmented Diffusion Estimation for Offline Black-Box Optimization
and 7 more in the interactive map.