ICLR 2025 accepted papers on Probabilistic and scientific modeling
288 of the 3,703 papers accepted at ICLR 2025 (7.8%) are on Probabilistic and scientific modeling, compared with 8.7% at ICLR 2024. They fall into 8 topics.
See ICLR 2025 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 · 71 papers
- Protein Language Model Fitness is a Matter of Preference (25 citations)
- IgGM: A Generative Model for Functional Antibody and Nanobody Design (24 citations)
- SynFlowNet: Design of Diverse and Novel Molecules with Synthesis Constraints (12 citations)
- ShEPhERD: Diffusing shape, electrostatics, and pharmacophores for bioisosteric drug design (7 citations)
- Generative Flows on Synthetic Pathway for Drug Design (5 citations)
- CBGBench: Fill in the Blank of Protein-Molecule Complex Binding Graph (4 citations)
- Structure Language Models for Protein Conformation Generation (4 citations)
- AtomSurf: Surface Representation for Learning on Protein Structures (3 citations)
- Fast Uncovering of Protein Sequence Diversity from Structure (3 citations)
- Hotspot-Driven Peptide Design via Multi-Fragment Autoregressive Extension (3 citations)
- MADGEN: Mass-Spec attends to De Novo Molecular generation (3 citations)
- On Conformal Isometry of Grid Cells: Learning Distance-Preserving Position Embedding (3 citations)
and 59 more in the interactive map.
Differential equations and PDEs · 42 papers
- ConFIG: Towards Conflict-free Training of Physics Informed Neural Networks (2 citations)
- Deep Learning Alternatives Of The Kolmogorov Superposition Theorem (2 citations)
- Active Learning for Neural PDE Solvers (1 citation)
- Air Quality Prediction with Physics-Guided Dual Neural ODEs in Open Systems (1 citation)
- On the Benefits of Memory for Modeling Time-Dependent PDEs (1 citation)
- Spectral-Refiner: Accurate Fine-Tuning of Spatiotemporal Fourier Neural Operator for Turbulent Flows (1 citation)
- Text2PDE: Latent Diffusion Models for Accessible Physics Simulation (1 citation)
- ANaGRAM: A Natural Gradient Relative to Adapted Model for efficient PINNs learning
- Efficient Training of Neural Stochastic Differential Equations by Matching Finite Dimensional Distributions
- Generalizable Motion Planning via Operator Learning
- Generating Physical Dynamics under Priors
- Graph Neural Preconditioners for Iterative Solutions of Sparse Linear Systems
and 30 more in the interactive map.
Variational and generative sampling · 40 papers
- Fast training and sampling of Restricted Boltzmann Machines (2 citations)
- PQMass: Probabilistic Assessment of the Quality of Generative Models using Probability Mass Estimation (2 citations)
- Deep MMD Gradient Flow without adversarial training (1 citation)
- Designing Mechanical Meta-Materials by Learning Equivariant Flows (1 citation)
- Multilevel Generative Samplers for Investigating Critical Phenomena (1 citation)
- Neural Sampling from Boltzmann Densities: Fisher-Rao Curves in the Wasserstein Geometry (1 citation)
- Restructuring Vector Quantization with the Rotation Trick (1 citation)
- Accelerating 3D Molecule Generation via Jointly Geometric Optimal Transport
- Continuous Autoregressive Modeling with Stochastic Monotonic Alignment for Speech Synthesis
- Controllable Generation via Locally Constrained Resampling
- Flow matching achieves almost minimax optimal convergence
- Flow Matching with General Discrete Paths: A Kinetic-Optimal Perspective
and 28 more in the interactive map.
Causal discovery · 38 papers
- Causal Order: The Key to Leveraging Imperfect Experts in Causal Inference (7 citations)
- Causal Representation Learning from Multimodal Biomedical Observations (1 citation)
- From Probability to Counterfactuals: the Increasing Complexity of Satisfiability in Pearl's Causal Hierarchy (1 citation)
- Unifying Causal Representation Learning with the Invariance Principle (1 citation)
- A Conditional Independence Test in the Presence of Discretization
- A Meta-Learning Approach to Bayesian Causal Discovery
- A Skewness-Based Criterion for Addressing Heteroscedastic Noise in Causal Discovery
- Causal Concept Graph Models: Beyond Causal Opacity in Deep Learning
- Causal Discovery via Bayesian Optimization
- Content-Style Learning from Unaligned Domains: Identifiability under Unknown Latent Dimensions
- Counterfactual Concept Bottleneck Models
- Counterfactual Generative Modeling with Variational Causal Inference
and 26 more in the interactive map.
Dynamical systems · 37 papers
- Boltzmann priors for Implicit Transfer Operators (2 citations)
- BRAID: Input-driven Nonlinear Dynamical Modeling of Neural-Behavioral Data (2 citations)
- Balanced Neural ODEs: nonlinear model order reduction and Koopman operator approximations (1 citation)
- Efficiently Parameterized Neural Metriplectic Systems (1 citation)
- Meta Flow Matching: Integrating Vector Fields on the Wasserstein Manifold (1 citation)
- SIMPL: Scalable and hassle-free optimisation of neural representations from behaviour (1 citation)
- A deep inverse-mapping model for a flapping robotic wing
- A Riemannian Framework for Learning Reduced-order Lagrangian Dynamics
- Artificial Kuramoto Oscillatory Neurons
- Continuity-Preserving Convolutional Autoencoders for Learning Continuous Latent Dynamical Models from Images
- CREIMBO: Cross-Regional Ensemble Interactions in Multi-view Brain Observations
- Discovering Temporally Compositional Neural Manifolds with Switching Infinite GPFA
and 25 more in the interactive map.
Uncertainty and Gaussian processes · 30 papers
- Scalable Bayesian Learning with posteriors (3 citations)
- From Risk to Uncertainty: Generating Predictive Uncertainty Measures via Bayesian Estimation (2 citations)
- Inverse decision-making using neural amortized Bayesian actors (1 citation)
- A Unified Theory of Quantum Neural Network Loss Landscapes
- Can Generative AI Solve Your In-Context Learning Problem? A Martingale Perspective
- CBMA: Improving Conformal Prediction through Bayesian Model Averaging
- Conformalized Interactive Imitation Learning: Handling Expert Shift and Intermittent Feedback
- Credal Wrapper of Model Averaging for Uncertainty Estimation in Classification
- Deep Kernel Posterior Learning under Infinite Variance Prior Weights
- Deep Random Features for Scalable Interpolation of Spatiotemporal Data
- Distribution-Free Data Uncertainty for Neural Network Regression
- ELBOing Stein: Variational Bayes with Stein Mixture Inference
and 18 more in the interactive map.
Treatment effects · 16 papers
- Interference Among First-Price Pacing Equilibria: A Bias and Variance Analysis (1 citation)
- Adaptive Shrinkage Estimation for Personalized Deep Kernel Regression in Modeling Brain Trajectories
- Comparing Targeting Strategies for Maximizing Social Welfare with Limited Resources
- Constructing Confidence Intervals for Average Treatment Effects from Multiple Datasets
- Do Contemporary Causal Inference Models Capture Real-World Heterogeneity? Findings from a Large-Scale Benchmark
- Doubly robust identification of treatment effects from multiple environments
- Incremental Causal Effect for Time to Treatment Initialization
- Learning Causal Alignment for Reliable Disease Diagnosis
- Model-agnostic meta-learners for estimating heterogeneous treatment effects over time
- Time After Time: Deep-Q Effect Estimation for Interventions on When and What to do
- Causal Effect Estimation with Mixed Latent Confounders and Post-treatment Variables
- Causal Graph Transformer for Treatment Effect Estimation Under Unknown Interference
and 4 more in the interactive map.
Bayesian optimization · 14 papers
- Bayesian Optimization of Antibodies Informed by a Generative Model of Evolving Sequences (3 citations)
- Standard Gaussian Process is All You Need for High-Dimensional Bayesian Optimization (2 citations)
- Optimistic Games for Combinatorial Bayesian Optimization with Application to Protein Design (1 citation)
- Optimizing Posterior Samples for Bayesian Optimization via Rootfinding (1 citation)
- Bayesian Optimization via Continual Variational Last Layer Training
- BOFormer: Learning to Solve Multi-Objective Bayesian Optimization via Non-Markovian RL
- Neural Exploratory Landscape Analysis for Meta-Black-Box-Optimization
- Offline Model-Based Optimization by Learning to Rank
- PABBO: Preferential Amortized Black-Box Optimization
- Sharpness-Aware Black-Box Optimization
- Brain Bandit: A Biologically Grounded Neural Network for Efficient Control of Exploration
- Gaussian Ensemble Belief Propagation for Efficient Inference in High-Dimensional, Black-box Systems
and 2 more in the interactive map.