Materials Discovery papers, 2023–2026

Key findings

• Materials Discovery went from 30 papers in 2023 to 77 in 2026, roughly 2.6x in three years.
• ICML publishes the most Materials Discovery work: 70 papers from 2023 to 2026, 34% of the four-venue total.
• The fastest-growing overlap is with Applications · chemistry physics and earth sciences: papers tagged with both rose from 1 in 2023 to 8 in 2026.
• Among the 5 topics under Climate and Physical Sciences, it ranks #3 by 2026 paper count; the fastest grower in the group, Weather Forecasting, is up 6.1x versus 2.6x here.

Representative papers

Spread across years and venues so every edition shows up

• NeurIPS 2026AMGenC: Generating Charge Balanced Amorphous Materials
• ICML 2026Adaptive Symmetry Discovery for Dynamical System Identification
• ICLR 2026Bures Generalized Category Discovery
• CVPR 2026TAR Token-Aware Refinement for Fine-grained Generalized Category Discovery
• NeurIPS 2025Generalized Category Discovery under Domain Shift: A Frequency Domain Perspective
• ICML 2025Neural Discovery in Mathematics: Do Machines Dream of Colored Planes?
• ICLR 2025DebGCD: Debiased Learning with Distribution Guidance for Generalized Category Discovery
• CVPR 2025Adaptive Part Learning for Fine-Grained Generalized Category Discovery: A Plug-and-Play Enhancement
• NeurIPS 2024WelQrate: Defining the Gold Standard in Small Molecule Drug Discovery Benchmarking
• ICML 2024Amortized Equation Discovery in Hybrid Dynamical Systems
• ICLR 2024Towards Faithful Explanations: Boosting Rationalization with Shortcuts Discovery
• CVPR 2024Eclipse: Disambiguating Illumination and Materials using Unintended Shadows
• NeurIPS 2023An Optimization-based Approach To Node Role Discovery in Networks: Approximating Equitable Partitions
• ICML 2023Deep Temporal Sets with Evidential Reinforced Attentions for Unique Behavioral Pattern Discovery
• ICLR 2023Improving the imputation of missing data with Markov Blanket discovery
• CVPR 2023Bootstrap Your Own Prior: Towards Distribution-Agnostic Novel Class Discovery
• NeurIPS 2026Ballad: Bandit-Based LLM Routing for Automated Heuristic Discovery
• ICML 2026CURE: Consistency-under-Unified Semantic Regularization for Generalized Category Discovery
• ICLR 2026From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers
• CVPR 2026Decouple Your Discovery and Memory in Continual Generalized Category Discovery

All Materials Discovery papers

33 papers, page 1 of 1

1. NeurIPS 2026Active Flow Expansion for Out-of-Distribution Discovery: from Theory to Molecules
2. NeurIPS 2026AI for Drug Discovery Models Often Do Not Learn as Expected and How to Diagnose These Failure Modes
3. NeurIPS 2026AIRA-Compose: Agentic Discovery of Neural Architectures
4. NeurIPS 2026AMGenC: Generating Charge Balanced Amorphous Materials
5. NeurIPS 2026Ballad: Bandit-Based LLM Routing for Automated Heuristic Discovery
6. NeurIPS 2026Causal Discovery from Unseen Environments
7. NeurIPS 2026Causal discovery needs explicit epistemic standards
8. NeurIPS 2026CausalTab: Pretraining Across Causal Environments for Tabular Causal Discovery
9. NeurIPS 2026Contrastive Discovery: Open-Ended Scientific Discovery over Competing Explanations
10. NeurIPS 2026Cross-Model Circuit Discovery
11. NeurIPS 2026Crosscoding Through Time: Sparse Feature Discovery Across Sequence Positions
12. NeurIPS 2026CruxBench: A Benchmark of Information Discovery
13. NeurIPS 2026Deep-Koopman-KANDy: Dictionary Discovery for Deep-Koopman Operators with Kolmogorov-Arnold Networks for Dynamics
14. NeurIPS 2026DrugSAGE: Self-evolving Agent Experience for Efficient State-of-the-Art Drug Discovery
15. NeurIPS 2026Fiedler-Regularized Causal Discovery for Sparse Connected DAGs
16. NeurIPS 2026Finite Resources False Discovery Rate Control on Structured Hypothesis Spaces
17. NeurIPS 2026From I/O to Code with Discovery Agent
18. NeurIPS 2026Hodge Laplacian Quasi-Harmonic Flows for Option Discovery
19. NeurIPS 2026HyperTree: Scalable Unsupervised Hierarchy Discovery in Hyperbolic Space
20. NeurIPS 2026Improving constraint-based discovery with robust propagation and LLM priors
21. NeurIPS 2026LEIA: Learned Environment for Interactive Architected Materials
22. NeurIPS 2026Mapping Uncharted Symmetries: Machine Discovery in Combinatorics
23. NeurIPS 2026MaterialsSaddles: 34 Million Transition States and a Flow-Matching Saddle-Point Predictor for Materials
24. NeurIPS 2026MOOD: Benchmarking Post-Hoc OOD Detection for Materials Property Prediction
25. NeurIPS 2026NORMA: Norm-Guided Explanation Subgraph Discovery
26. NeurIPS 2026Offline Materials Optimization with CliqueFlowmer
27. NeurIPS 2026PAIR-CI: Calibrated Conditional Independence Testing for Causal Discovery with Incomplete Data
28. NeurIPS 2026Preserving DEG Rankings for Gene Discovery in Histology-Based Spatial Gene Expression Prediction
29. NeurIPS 2026SciResearchBench: Benchmarking AI Agents on Complex Scientific Literature Discovery
30. NeurIPS 2026Seen-Constrained Model-Order Selection for Unknown-$K$ Generalized Category Discovery
31. NeurIPS 2026Self-Supervised Keyframe Discovery for Horizon-Invariant Behavior Cloning
32. NeurIPS 2026TagBO: LLM-Driven Task-Aware Graph Bayesian Optimization for Scientific Discovery
33. NeurIPS 2026Weather-Robust Cross-View Geo-Localization via Prototype-Based Semantic Part Discovery

FAQ

How many Materials Discovery papers were published at NeurIPS, ICLR, ICML and CVPR from 2023 to 2026?

208 in total: 30 in 2023, 48 in 2024, 53 in 2025, 77 in 2026.

How many Materials Discovery papers are at NeurIPS 2026?

NeurIPS 2026 has 33 Materials Discovery papers, versus 15 the year before.

How many Materials Discovery papers are at ICLR 2026?

ICLR 2026 has 11 Materials Discovery papers, versus 12 the year before.

How many Materials Discovery papers are at ICML 2026?

ICML 2026 has 27 Materials Discovery papers, versus 13 the year before.

How many Materials Discovery papers are at CVPR 2026?

CVPR 2026 has 6 Materials Discovery papers, versus 13 the year before.

Which conference publishes the most Materials Discovery papers?

ICML, with 70 papers from 2023 to 2026.