AI4Science News
Industry updates, funding rounds, and notable releases
Auto-updated from RSS feeds · Last sync: 2/3/2026
Measurement for Opaque Systems: Multi-source Triangulation with Interpretable Machine Learning
arXiv:2602.00022v1 Announce Type: new Abstract: We propose a measurement framework for difficult-to-access contexts that uses indirect data traces, interpretable machine-learning models, and theory...
Representation Learning Enhanced Deep Reinforcement Learning for Optimal Operation of Hydrogen-based Multi-Energy Systems
arXiv:2602.00027v1 Announce Type: new Abstract: Hydrogen-based multi-energy systems (HMES) have emerged as a promising low-carbon and energy-efficient solution, as it can enable the coordinated ope...
Sheaf Neural Networks and biomedical applications
arXiv:2602.00159v1 Announce Type: new Abstract: The purpose of this paper is to elucidate the theory and mathematical modelling behind the sheaf neural network (SNN) algorithm and then show how SNN...
GRIP2: A Robust and Powerful Deep Knockoff Method for Feature Selection
arXiv:2602.00218v1 Announce Type: new Abstract: Identifying truly predictive covariates while strictly controlling false discoveries remains a fundamental challenge in nonlinear, highly correlated,...
In-Run Data Shapley for Adam Optimizer
arXiv:2602.00329v1 Announce Type: new Abstract: Reliable data attribution is essential for mitigating bias and reducing computational waste in modern machine learning, with the Shapley value servin...
Prototype-based Explainable Neural Networks with Channel-specific Reasoning for Geospatial Learning Tasks
arXiv:2602.00331v1 Announce Type: new Abstract: Explainable AI (XAI) is essential for understanding machine learning (ML) decision-making and ensuring model trustworthiness in scientific applicatio...
MATRIX: A Multimodal Benchmark and Post-Training Framework for Materials Science
arXiv:2602.00376v1 Announce Type: new Abstract: Scientific reasoning in materials science requires integrating multimodal experimental evidence with underlying physical theory. Existing benchmarks ...
Open Materials Generation with Inference-Time Reinforcement Learning
arXiv:2602.00424v1 Announce Type: new Abstract: Continuous-time generative models for crystalline materials enable inverse materials design by learning to predict stable crystal structures, but inc...
AREAL-DTA: Dynamic Tree Attention for Efficient Reinforcement Learning of Large Language Models
arXiv:2602.00482v1 Announce Type: new Abstract: Reinforcement learning (RL) based post-training for large language models (LLMs) is computationally expensive, as it generates many rollout sequences...
NEST: Nested Event Stream Transformer for Sequences of Multisets
arXiv:2602.00520v1 Announce Type: new Abstract: Event stream data often exhibit hierarchical structure in which multiple events co-occur, resulting in a sequence of multisets (i.e., bags of events)...
Google DeepMind releases AlphaFold 3 for academic use
AlphaFold 3 model weights now available for academic research, expanding access beyond the AlphaFold Server.
NVIDIA launches BioNeMo 2.0 with new foundation models
Updated platform includes ESM-2, MegaMolBART, and new tools for molecular generation and optimization.
OpenAI partners with Moderna on mRNA design
ChatGPT Enterprise deployed for vaccine development workflows, marking major pharma AI partnership.
Chai Discovery raises $30M to challenge AlphaFold
Startup launches with open model claiming competitive performance on drug-target prediction benchmarks.
Schrödinger reports positive Phase 2 results for AI-designed drug
SGR-1505 shows promising efficacy in B-cell malignancies, validating computational drug design approach.
Recursion acquires Exscientia in $688M all-stock deal
Major consolidation in AI drug discovery as Recursion absorbs Exscientia, combining their platforms and pipelines.