[CLSTR]
< Back to situations

Monitor this situation.

[SITUATION] · [QUIET] · [TECHNOLOGY]

2 clusters · 4 sources · 20 days · First seen · Last updated

AI advancements in biological and physical modeling

Overview

Developments in artificial intelligence are increasingly focused on creating models capable of understanding complex biological and physical systems.

Initial research introduced a protein language model that integrates amino acid sequences with three-dimensional structures. This model allows scientists to create a ‘protein world map’ to study evolutionary patterns and relationships across the protein universe.

Subsequent advancements have expanded into the concept of ‘world models.’ In robotics, new bilinear world models have been proposed to accelerate planning by reducing time requirements while maintaining control accuracy. Simultaneously, in biological research, multimodal world models like Quine have been introduced to integrate diverse data types—including genomics, proteins, and bioimaging—to predict how biological systems respond to specific interventions.

Entities

University of Haifa · Institute of Science Tokyo · Quine · Tel Aviv University · Microsoft Research

Timeline

  1. [TECHNOLOGY] 2 sources
    AI world models advance robotics planning and biological research

    New AI world models are advancing robotics through faster planning via bilinear parameterization and biology through Microsoft's Quine, which integrates multi-scale biological data to predict cellular responses

  2. [TECHNOLOGY] 2 sources
    AI model integrates protein sequences and structures to map evolution

    Researchers have developed an AI protein language model that combines amino acid sequences and 3D structures to map protein evolution and relationships across the protein universe.

Sources

azolifesciences.com · donanimhaber.com · isaret.ai · news-medical.net