-
lamm-mit/gemma4-jacobian-lenses
Updated • 1 -
lamm-mit/gemma4-materials-latent-vectors
Updated • 66 -
lamm-mit/gemma4-materials-mechanism-prompts
Viewer • Updated • 3.05k • 100 -
Reading and Steering Representations of Materials-Science Mechanisms in an Open-Weight Language Model
Paper • 2607.20058 • Published • 5
AI & ML interests
PI: Markus J. Buehler, MIT. Our research focus on developing a new paradigm that designs materials from the molecular scale, using MD, ML and other methods.
Recent Activity
Papers
Reading and Steering Representations of Materials-Science Mechanisms in an Open-Weight Language Model
Autonomous Agents Coordinating Distributed Discovery Through Emergent Artifact Exchange
The Laboratory for Atomistic and Molecular Mechanics (LAMM) at MIT develops AI for science methods to model, reason about, and discover new materials and biological systems.
Our work combines machine learning, graph reasoning, generative models, autonomous agents, molecular simulation, and multiscale mechanics to accelerate scientific discovery across materials science, biology, and engineering.
We share models, datasets, and tools for AI-driven scientific reasoning, materials design, and molecular-scale discovery.
PI: Prof. Markus J. Buehler, MIT, mbuehler@MIT.EDU Lab: Laboratory for Atomistic and Molecular Mechanics
-
lamm-mit/gemma4-jacobian-lenses
Updated • 1 -
lamm-mit/gemma4-materials-latent-vectors
Updated • 66 -
lamm-mit/gemma4-materials-mechanism-prompts
Viewer • Updated • 3.05k • 100 -
Reading and Steering Representations of Materials-Science Mechanisms in an Open-Weight Language Model
Paper • 2607.20058 • Published • 5
spaces 6
Graphene Design Universe 256K
Explore 256,000 graphene designs with atomistic coordinates
Graphene Design Universe
Explore 64,000 graphene designs with atomistic coordinates
PDF2Audio
Generate spoken‑ready scripts from documents
Graph-PrefLexOR Scientific Reasoning
Graph-native LLM scientific reasoning with graph viz
TRELLIS.2
High-fidelity 3D Generation from images