NeurIPS 2025 Past AI for science
NeurIPS 2025 AI for Science Workshop
NeurIPS2025-AI4Science
- Submission deadline
- Aug 28, 2025, 11:59 UTC imported from OpenReview — check the website for extensions
- Submission portal
- OpenReview
- Notes
- Auto-imported from the OpenReview venue record on 2026-06-10 — please verify and enrich (topics are keyword-guessed).
Accepted papers (230)
Fetched from OpenReview (v2) on 2026-06-10.
-
10 Million Particle Events: Enabling Foundation Models for Sparse 3D Inverse Problems
-
A Foundational Dataset for the Predictive Prevention of Waterborne Disease
-
A Large Multimodal Molecular Representation Encoder-Decoder Foundation Model for Chemistry
-
A Multi-Modal Deep Learning Model for Drug Potency Prediction: Leveraging Features from Physics-Based Docking and Advanced Co-Folding Methods
-
A Probabilistic U-Net Approach to Downscaling Climate Simulations
-
A study of EHVI vs fixed scalarization for molecule design
-
A Synthesizability-Guided Pipeline for Materials Discovery
-
AC-PKAN: Attention-Enhanced and Chebyshev Polynomial-Based Physics-Informed Kolmogorov–Arnold Networks
-
Accelerated Isotopologue Reduced Partition Function Ratio Prediction with Orbital-based Deep Learning
-
Accelerating Protein Molecular Dynamics Simulation with DeepJump
-
Adaptive Transition State Refinement with Learned Equilibrium Flows
-
AI for Science Strategic Compass: Aligning Discovery Tensions with Core AI Functions
-
AI4O3: A Foundational Data Collection for Artificial Intelligence in Tropospheric Ozone Research
-
AIM: Adaptive Intervention for Deep Multi-task Learning of Molecular Properties
-
AION-1: Omnimodal Foundation Model for Astronomical Sciences
-
Alvessa: An Agentic Evidence-Grounded Research Assistant for Genomics
-
An Agentic Orchestration System for Heliophysics Tasks
-
An in-silico integration of neurodevelopmental and dopaminergic views of schizophrenia
-
Assessing the Geographic Generalization and Physical Consistency of Generative Models for Climate Downscaling
-
Augmenting Research Ideation with Data: An Empirical Investigation in Social Science
-
AutoChemSchematic AI: Agentic Physics-Aware Automation for Chemical Manufacturing Scale-Up
-
Automated scientific minimization of regret for cognitive modeling
-
BasePrompt: Self-Prompting Genome Language Models for RNA Fitness Prediction
-
Benchmarking LLMs for atomic-level geometric manipulation in crystals
-
Benchmarking Machine Learning Potentials for Crystal Structure Relaxation
-
Beyond Atoms: Evaluating Electron Density Representation for 3D Molecular Learning
-
Beyond data subsampling: differentiation as an uncertainty source in equation discovery
-
Beyond Ensembles: Simulating All-Atom Protein Dynamics in a Learned Latent Space
-
Beyond model organisms: robust prediction of functional properties across protein evolution
-
Bigger is not always better: evaluating target-specific dataset design strategies for regioselectivity prediction on complex molecules
-
BioMedReasoner: Towards Multi-Hop Reasoning using Path-based Relational Learning on Biomedical Knowledge Graphs
-
BioVerge: A Comprehensive Benchmark and Study of Self-Evaluating Agents for Biomedical Hypothesis Generation
-
Block-wise distillation for lightweight weather models
-
BLOSUM Is All You Learn — Generative Antibody Models Reflect Evolutionary Priors
-
Boundary-Augmented Neural Operators for Better Generalization to Unseen Geometries
-
Bridging Neural Operator and Flow Matching for a Generative PDE Foundation Model
-
CALM-PDE: Continuous and Adaptive Convolutions for Latent Space Modeling of Time-dependent PDEs
-
Can Theoretical Physics Research Benefit from Language Agents?
-
CAST: Causal Modeling of Time-Varying Treatment Effects on Head and Neck Cancer
-
Causal AI Scientist: Facilitating Causal Data Science with Large Language Models
-
Chemist-aligned retrosynthesis by ensembling diverse inductive bias models
-
CHEMSETS: How Capable Are Chemistry LLMs?
-
CiteGuard: Retrieval-Augmented Citation Verification for LLM-Powered Peer Review
-
Closing the Omics Gap: A Benchmark for Unified Evaluation of Biomolecular Foundation Models
-
CompGen: A Conditional Generation Framework for Inverse Composition Design of Catalytic Surfaces
-
Conditioned Clifford-Steerable Kernels
-
Connecting Preclinical Models to Patient Outcomes: A Machine Learning Dataset for Predictive Validity in Drug Development
-
Consistent Synthetic Sequences Unlock Structural Diversity in Fully Atomistic De Novo Protein Design
-
Constant-Potential Machine Learning Force Field for Electrochemical Interface
-
Constructing the Mental Health Phenome: An Open Multimodal Dataset Linking Digital Behavior, Physical Health, and Mental Wellbeing
-
Control-Augmented Diffusion for Autoregressive Data Assimilation
-
Data-Dependent Smoothing for Protein Discovery with Walk-Jump Sampling
-
Data-driven Design as a High-Impact, Ecologically Valid Benchmark for Document Understanding
-
Data-Driven Solar Surface Flux Transport Modeling with Uncertainty Quantification
-
Data-optimal scaling of paired antibody language models
-
De novo generation of functional terpene synthases using TpsGPT
-
Decompose, Adapt, and Evolve: Towards Efficient Scientific Equation Discovery with Large Language Models
-
Deep Graph Learning for Industrial Carbon Emission Analysis and Policy Impact
-
Demystifying Protein Generation with Hierarchical Conditional Diffusion Models
-
Differentiable Predictive Control for Precise Oxygen Level Maintenance for Critical Patients
-
Diffusion for Fusion: Designing Stellarators with Generative AI
-
Dimensionality and Topological Stability of Neural Representations in the Human Brain Predict Learning Outcomes
-
DINO: dynamics-informed dataset to overcome the limitations of static molecular data in AI-driven drug discovery
-
Discontinuous Epitope Fragments as Sufficient Target Templates for Efficient Binder Design
-
Dissecting Larval Zebrafish Hunting Behavior using Deep Reinforcement Learning trained RNNs
-
DistMLIP: A Distributed Inference Platform for Machine Learning Interatomic Potentials
-
Diverse Topology Optimization using Modulated Neural Fields
-
DMPKBench: A Multi-Modal Benchmark for Evaluating LLMs and Agents in Drug Discovery DMPK Tasks
-
DMRG Quantum Chemistry Dataset for Multi-Reference Machine Learning
-
Do Llamas Understand the Periodic Table?
-
Does LLM dream of differential equation discovery?
-
Domain-Invariant Feature Learning for Patient-Level Phenotype Prediction from Single-Cell Data
-
EARS-UDE : Evaluating Auditory Response in Sensory Overload with Universal Differential Equations
-
Einstein Fields: A Neural Perspective To Computational General Relativity
-
Emergent SO(3)-Invariant Molecular Representations from Multimodal Alignment
-
Empowering AI in RNAi Therapeutics: A Foundational Dataset for siRNA Design and Optimization
-
EquiHGNN: Scalable Rotationally Equivariant Hypergraph Neural Networks
-
Every Answer Counts: Enhancing Scientific Discovery with Efficient Entity-Centric Question Answering from Long Contexts
-
Explainable AI–Guided Virtual Experiments Reveal How DNA Sequence Context Shapes Gene Regulation
-
Explaining Temporal Effects in Sepsis Prediction
-
Exploring Generative Approaches for Predicting Copolymer Sequences from Reaction Conditions
-
FALCON: An ML Framework for Fully Automated Layout-Constrained Analog Circuit Design
-
Few-shot Protein Fitness Prediction via In-context Learning and Test-time Training
-
First Comprehensive Benchmark for Tailored Small Molecule-Binding Aptamer Design
-
Foundation Models Enabling Multi-Scale Battery Materials Discovery: From Molecules To Devices
-
From In Silico to In Vitro: Evaluating Molecule Generative Models for Hit Generation
-
From Molecules to Perception: A Benchmark Dataset for AI in Sensory Science
-
From Static Structures to Ensembles: Studying and Harnessing Protein Structure Tokenization
-
GAPMAP: Mapping Scientific Knowledge Gaps in Biomedical Literature Using Large Language Models
-
GCP-VQVAE: A Geometry-Complete Language for Protein 3D Structure
-
Generalization Beyond Benchmarks: Evaluating Learnable Protein-Ligand Scoring Functions on Unseen Targets
-
Generative AI Enables Medical Image Segmentation in Ultra Low-Data Regimes
-
Generative Latent Space Dynamics of Electron Density
-
GeoGraph: Geometric and Graph-based Ensemble Descriptors for Intrinsically Disordered Proteins
-
Geometry Aware Inference of Steady State PDEs Using Equivariant Neural Field Representations
-
Gradient-Free Physics-informed Operator Learning using Walk-on-Spheres
-
Graph Neural Networks for Interferometer Simulations
-
Hash Collisions in Molecular Fingerprints: Effects on Property Prediction and Bayesian Optimization
-
Holonic Science: A New Framework for Benchmarking AI Scientists
-
How knowledge discovery and embedded paradigm transform industrial process management: exploring pipeline hydraulic dynamic identification
-
How to Detect and Defeat Molecular Mirage: A Metric-Driven Benchmark for Hallucination in LLM-based Molecular Comprehension
-
IM-LPG: Inverse Modeling Approach to Laser Pulse Shape Generation in Inertial Confinement Fusion
-
Improved Therapeutic Antibody Reformatting through Multimodal Machine Learning
-
Improving RNA Secondary Structure Prediction Through Expanded Training Data
-
Is Sequence Information All You Need for Bayesian Optimization of Antibodies?
-
Label-free biochemical imaging of neural organoids via deep learning-enhanced Raman microspectroscopy
-
Large-scale audio-language datasets for bioacoustics
-
LeafTrackNet: A Deep Learning Framework for Robust Leaf Tracking in Top-Down Plant Phenotyping
-
Learning Boltzmann Generators via Constrained Mass Transport
-
Learning chaotic PDEs with boundedness guarantees
-
Learning Deformable Body Interactions With Adaptive Spatial Tokenization
-
Learning Protein-Ligand Binding in Hyperbolic Space
-
Learning to Compress Plasma Turbulence
-
LEONARDO: A Physics-Informed Generative Model for Stochastic Nanoparticle Dynamics in Liquid-Phase TEM
-
Leveraging Chemistry Foundation Models to Facilitate Structure Focused Retrieval Augmented Generation in Multi-Agent Workflows for Catalyst and Materials Design
-
LINKER: Learning Interactions Between Functional Groups and Residues With Chemical Knowledge-Enhanced Reasoning and Explainability
-
LLM Kernel: an evaluation framework for open-ended scientific interpretation
-
Machine Learning Interatomic Potentials: library for efficient training, model development and simulation of molecular systems
-
MARSHA: Multi-Agent RAG System for Hazard Adaptation
-
Measuring Dependencies between Biological Signals with Self-supervision, and its Limitations
-
Mechanistic Reaction Data for Interpretable Deep Learning in Chemistry
-
MEGA: A Large-Scale Molecular Editing Dataset for Guided-Action Optimization
-
Memory-Augmented Reinforcement Learning for Hierarchical Graph Optimization of Dynamic Bills of Materials in Sustainable Medical device Product Families
-
MetaOmics-10T: The Foundational Dataset to Unlock Causal Modeling of Microbial Ecosystems
-
Mixture-of-Experts Guided Multi-Omic Integration for Gastrointestinal Cancer Subtype Prediction
-
MLIPAudit: A benchmarking tool for Machine Learned Interatomic Potentials
-
Mol-LLaMA: Towards General Understanding of Molecules in Large Molecular Language Model
-
Mol-LLM: Multimodal Generalist Molecular LLM with Improved Graph Utilization
-
Mol-SGCL: Molecular Substructure-Guided Contrastive Learning for Out-of-Distribution Generalization
-
MOOSE-Chem2: Exploring LLM Limits in Fine-Grained Scientific Hypothesis Discovery \\via Hierarchical Search
-
MOOSE-Chem3: Toward Experiment-Guided Hypothesis Ranking via Simulated Experimental Feedback
-
moPPIt-v3: Motif-Specific Peptides Generated via Multi-Objective-Guided Discrete Flow Matching
-
MORGaN: self-supervised multi-relational graph learning for drug target discovery
-
MSAFlow: a Unified Approach for MSA Representation, Augmentation, and Family-based Protein Design
-
Multi-Graph Meta-Transformer: An Interpretable Framework for Cross-Graph Functional Alignment in Neural Decoding
-
Multi-Modal Attention Framework for Underwater Bioacoustic Denoising and Recognition
-
Multi-Objective Nanobody Design via Masked Discrete Diffusion with Simplex Refinement
-
Multi-Objective Peptide Design via Token-Aligned Preference Optimization
-
Multi-Scale Classification of Green Bank Telescope Signals
-
Multilevel neural simulation-based inference
-
Multimodal Large Language Models for Inverse Molecular Design with Retrosynthetic Planning
-
Multiscale Neural PDE Surrogates for Prediction and Downscaling: Application to Ocean Currents
-
Neural network distillation of orbital dependent density functional theory
-
Neural Triangular Transport Maps: A New Approach Towards Sampling in Lattice QCD
-
OmniCast: A Masked Latent Diffusion Model for Weather Forecasting Across Time Scales
-
OpenCityCorpus: A Large-Scale, Harmonized, and LLM-Ready Corpus of Urban Data for Scientific Research
-
OpenDiscovery: A Verifiable, Creative Science Problem-Solving Dataset to Forge AI Scientists
-
Pareto-Guided Reinforcement Learning for Multi-Objective ADMET Optimization in Generative Drug Design
-
PatchDNA: A Flexible and Biologically-Informed Alternative to Tokenization for DNA
-
PEAR: Equal Area Weather Forecasting on the Sphere
-
PepThink-R1: LLM for Interpretable Cyclic Peptide Optimization with CoT SFT and Reinforcement Learning
-
Perovskite-LLM: Knowledge-Enhanced Large Language Models for Perovskite Solar Cell Research
-
PhySense: Evaluating LLMs on Foundational Physics Principles
-
Physics-Informed Learning Near Critical Transitions: A Comparative Study of UDEs and Neural ODEs
-
Physics-Informed Neural Networks with Fourier Features and Attention-Driven Decoding
-
PhysiX: A Foundation Model for Physics Simulations
-
PICore: Physics-Informed Unsupervised Coreset Selection for Data Efficient Neural Operator Training
-
PIRF: Physics-Informed Reward Fine-Tuning for Diffusion Models
-
PKG-DPO: Optimizing Domain-Specific AI systems with Physics Knowledge Graphs and Direct Preference Optimization
-
PLAME: Lightweight MSA Design Advances Protein Folding From Evolutionary Embeddings
-
Predicting Kinase-Specific Phosphorylation Sites with Pretrained Protein Language Models
-
Predictive Feature Caching for Training-free Acceleration of Molecular Geometry Generation
-
PrimerCast: Predictive Modeling of PCR Amplification with an AI-Ready Experimental Dataset
-
Proposal for a Large-scale High-quality Dataset of Activity Cliffs
-
Protein Design with Agent Rosetta: A Case Study for Specialized Scientific Agents
-
PUBHOMICS: A Multispecies Biological Dataset to Catalyze AI-Driven Toxicity Assessment for Environmental and Public Health
-
RAG-Enhanced Collaborative LLM Agents for Drug Discovery
-
Rao-Blackwell Gradient Estimators for Equivariant Denoising Diffusion
-
ReactionReasoner: Towards Reasoning LLM for Chemical Reaction Prediction
-
README: Rapid Equation Discovery with Multimodel Encoders
-
Reasoning LLMs for Materials Discovery with Physics-aware Rejection Sampling
-
RemoteFoldSet: Benchmarking Structural Awareness of Protein Language Models
-
Resilience Outcomes Benchmark: Toward an Outcome-Labeled Coping Strategy Dataset for Precision Mental Health
-
Reviewing Scientific Papers for Critical Problems With Reasoning LLMs: Baseline Approaches and Automatic Evaluation
-
Revive Legacy Scientific Reasoning Benchmark by Growing Perturbation
-
RNA-Scope: Benchmarking RNA Language Models for RNA Sequence Understanding
-
Rodent-Bench
-
SafeScientist: Toward Risk-Aware Scientific Discoveries by LLM Agents
-
Sampling 3D Molecular Conformers with Diffusion Transformers
-
Scaling High-Throughput Experimentation Unlocks Robust Reaction-Outcome Prediction
-
Scaling Multi-Modal and Multi-Task Transformers for Small Molecule Drug Discovery
-
scCMap: Connecting Genetic and Chemical Perturbations at Single-Cell Resolution
-
Scientific Machine Learning for Symbolic Recovery of Relativistic Effects in Black Hole Orbits
-
SciKnowEval: A Comprehensive Dataset for Evaluating Scientific Knowledge of Large Language Models
-
SciNav: A General Agent Framework for Scientific Coding Tasks
-
Semantic search for 100M+ galaxy images using AI-generated captions
-
Sinhala Diachronic Corpus
-
SkillPuzzler: A Self-Evolving Agentic Framework for Materials and Chemistry Research with Minimal Reliance on Predefined Tools
-
Smiles2Dock: a large-scale dataset for ML-based docking score prediction using AlphaFold structures
-
Softly Constrained Denoisers for Diffusion Models
-
SPADE: Inferring Transcriptional Dynamics from Spatial Transcriptomics with Physics-Informed Deep Learning
-
Sparse Autoencoders for Low-$N$ Protein Function Prediction and Design
-
Sparse Mixture-of-Experts for Multi-Channel Imaging: Are All Channel Interactions Required?
-
Spatio-Temporal Graphs Beyond Grids: Benchmark for Maritime Anomaly Detection
-
Static and Dynamic Diffusion Emulators: From Sampling Gray Swan Extreme Events to Suffering from Model Collapse
-
Steering the Evolutionary Game: Hierarchical Control of Therapeutic Resistance in Cancer Treatment
-
Steering Vector Fields for Property-Controlled Molecular Generation with Chemical Language Models
-
Synergizing Large Language Models and Knowledge Graphs in Science: A Survey
-
SynthFair: A Semi-Synthetic Medical Imaging Dataset to Propel Research on Bias Detection & Mitigation
-
TadABench-1M: A Large-Scale Wet-Lab Protein Benchmark For Rigorous OOD Evaluation
-
Task Alignment Outweighs Framework Choice in Scientific LLM Agents
-
TeBaAb: Text-Based Antigen-Conditioned Antibody Redesign via Directed Evolution
-
Test-Time Control Over Accuracy-Cost Trade-Offs in Neural Physics Simulators via Recurrent Depth
-
The Darwin–Gödel Discovery Machine: Toward Bounded-Risk Self-Improving AI4Science
-
The Loss Landscape of XRD-Based Structure Optimization Is Too Rough for Gradient Descent
-
The More You Automate, the Less You See: The Hidden Pitfalls of AI Scientist Systems
-
The Transparent Earth: A Multimodal Foundation Model for the Earth's Subsurface
-
Thinking like a CHEMIST: Combined Heterogeneous Embedding Model Integrating Structure and Tokens
-
Token-Level Early Fusion Model Bridging Text and 3D Electron Density Grids in Chemistry
-
Token-Level Guided Discrete Diffusion for Membrane Protein Design
-
Topological defects propagate information in deep neural networks
-
Topological Feature Compression for Molecular Graph Neural Networks
-
Topological Graph Generative Model for Ecological Design
-
TorchQuantumDistributed
-
Towards Accurate Test-Time Adaptation for Neural Surrogates
-
Towards Generating Stable Materials via Large Language Models with Reinforcement Learning Finetuning
-
Towards Multi-Fidelity Scaling Laws of Neural Surrogates in CFD
-
Training Dynamics of Learning 3D-Rotational Equivariance
-
TroubleRAG: Evaluating Retrieval Pipelines for Real-World Chemistry Troubleshooting
-
Trustworthy Retrosynthesis: Eliminating Hallucinations with a Diverse Ensemble of Reaction Scorers
-
Unlearning as Ablation: Toward a Falsifiable Benchmark for Generative Scientific Discovery
-
Urban Climate Counterfactuals: A Causal Dataset for Street-Level Heat Mitigation Interventions
-
Using Deep Reinforcement Learning to Understand Odor Plume Tracking in Walking and Flying Agents
-
WhaleLM: Finding Structure and Information in Sperm Whale Vocalizations and Behavior with Machine Learning
-
When Do LLMs Improve Bayesian Optimization? A Systematic Comparison Across Molecular and Protein Design
-
WildSci: Advancing Scientific Reasoning from In-the-Wild Literature
-
Without Safeguards, AI-Biology Integration Risks Accelerating Future Pandemics
-
Wrong Model, Right Uncertainty: Spatial Associations for Discrete Data with Misspecification
-
Zephyrus: An Agentic Framework for Weather Science
-
Zero-Shot Protein–Ligand Binding-Residue Prediction from Sequence and SMILES