materials science
PulseAugur coverage of materials science — every cluster mentioning materials science across labs, papers, and developer communities, ranked by signal.
5 day(s) with sentiment data
-
New AI Model AIDEN Accelerates Materials Science Calculations
Researchers have developed AIDEN, an Atomic-Interaction Density Equivariant Network, designed to solve real-space charge density problems. This deep learning model separates element-dependent densities from environmenta…
-
AI's potential for bioweapon creation demands close monitoring of its scientific abilities
The potential for AI to engineer dangerous viruses is a significant short-term risk that warrants close monitoring. While highly advanced AI could orchestrate such an attack, the immediate concern lies with more moderat…
-
AI model infers material microstructures from X-ray diffraction data
Researchers have developed a novel cross-modal learning framework to infer 3D dislocation microstructures directly from X-ray diffraction data. This approach embeds representations of dislocation density fields and thei…
-
AI, Quantum Computing, and Autonomous Labs Converge to Reshape Chemistry
A new commentary piece titled "The convergent laboratory" explores the transformative impact of AI, autonomous agents, high-performance computing, and quantum computing on chemistry and materials science. The authors, d…
-
New COGTRL framework trains LLMs for scientific discovery with cognitive traces
Researchers have developed COGTRL, a novel reinforcement learning framework designed to enhance the capabilities of large language models (LLMs) as scientific discovery assistants. By training LLMs to generate "cognitiv…
-
New research explores adaptable and domain-independent neural operators · 4 sources tracked
Researchers are exploring new methods for neural operators, which are used to approximate physical simulations. One approach, LatentDDM, focuses on pretraining operators on smaller subdomains and then using a lightweigh…
-
Materials science ontology paper reaches 100 citations
A paper titled "PMD Core Ontology: Achieving semantic interoperability in materials science" has surpassed 100 citations. This mid-level ontology is designed to enhance semantic interoperability within materials science…
-
Machine learning models for materials discovery face critical design flaw
A new research paper published on arXiv highlights a critical design choice in machine learning models used for materials discovery. The study demonstrates that whether a model predicts physically impossible properties,…
-
LLMs poised to revolutionize nanophotonics design and discovery
A new review paper explores the integration of Large Language Models (LLMs) into the field of nanophotonics. The paper details how LLMs are moving beyond traditional neural networks by providing semantic interfaces, gen…
-
Foundation models and LLMs advance nanophotonic design and discovery
Researchers have developed MOCLIP, a foundation model for nanophotonic inverse design, leveraging contrastive learning to integrate geometry and spectral representations. This model achieves high-throughput zero-shot pr…
-
New benchmark reveals multimodal LLMs struggle with scientific discovery
A new benchmark called Science Edge Evaluation (SEE) has been developed to assess the capabilities of multimodal large language models (MLLMs) in complex scientific discovery tasks. Across 19 MLLMs, the highest accuracy…
-
Meta AI models power Genesis Mission projects at Lawrence Berkeley Lab
Meta's AI models, including the Segment Anything Model and those built on PyTorch, are being utilized in the initial projects of the Genesis Mission at Lawrence Berkeley National Laboratory. These projects focus on area…
-
Materials science innovation essential for next-gen AI infrastructure
Advanced materials are becoming crucial for the continued innovation in artificial intelligence, as current AI technologies push the limits of semiconductors and data centers. Innovations in polymers, specialty fluids, …
-
Lila Sciences builds AI-powered automated lab for scientific discovery
Lila Sciences is building an automated laboratory designed to function as an AI data center, generating vast amounts of experimentally validated scientific data. Their vision is to create a scientific superintelligence …
-
New hybrid method enhances neural operators for complex multiscale problems
Researchers have developed LOD-MSNO, a novel hybrid approach that combines the LOD method with neural operators to address challenges in solving multiscale problems. This method aims to improve the accuracy of neural op…
-
New benchmark reveals VLM limitations in materials science phase diagram understanding
Researchers have introduced MatPhaseBench, a new benchmark designed to evaluate the capabilities of Vision-Language Models (VLMs) in understanding complex materials science phase diagrams. This benchmark, derived from s…
-
New AI model enhances scientific hypothesis generation with traceable reasoning
Researchers have developed Graph-PRefLexOR, a novel graph-native reinforcement learning model designed to enhance scientific hypothesis generation. This model, fine-tuned using Group Relative Policy Optimization (GRPO),…
-
Materials Science Milestone Paves Way for Next-Gen Tech
A significant milestone has been achieved in materials science and engineering, paving the way for the next generation of technology. This advancement is expected to accelerate the mass production of materials crucial f…
-
Diffusion Model Predicts Crystal Structures from X-Ray Diffraction Data
Researchers have developed XRDiff, a novel diffusion model capable of predicting crystal structures from powder X-ray diffraction (PXRD) data. This model can infer structures based on known stoichiometry or, more challe…
-
AI framework improves defect classification in materials science imaging
Researchers have developed a context-aware deep learning framework to improve defect classification in atomic-resolution STEM imaging. This new approach integrates image contrast with metadata such as composition and be…