MIT’s CRESt platform links multimodal science data to drive materials discovery
MIT researchers launched **CRESt**, a multimodal AI platform that integrates data from scientific texts, images, tables, and experiments. CRESt autonomously designs and runs lab experiments to discover new materials — for instance, predicting stable compounds or suggesting synthesis routes. It learns from rich multimodal sources and closes the loop by validating in real labs. This system promises to accelerate materials innovation by uniting literature, simulations, and lab feedback in one AI pipeline.
positive
2 days ago
MIT’s CRESt platform links multimodal science data to drive materials discovery
MIT researchers launched **CRESt**, a multimodal AI platform that integrates data from scientific texts, images, tables, and experiments. CRESt autonomously designs and runs lab experiments to discover new materials — for instance, predicting stable compounds or suggesting synthesis routes. It learns from rich multimodal sources and closes the loop by validating in real labs. This system promises to accelerate materials innovation by uniting literature, simulations, and lab feedback in one AI pipeline.
positive
MIT’s CRESt platform links multimodal science data to drive materials discovery
2 days ago
1 min read
72 words
MIT’s CRESt uses multimodal AI to automate discovery of novel materials.
MIT researchers launched **CRESt**, a multimodal AI platform that integrates data from scientific texts, images, tables, and experiments. CRESt autonomously designs and runs lab experiments to discover new materials — for instance, predicting stable compounds or suggesting synthesis routes. It learns from rich multimodal sources and closes the loop by validating in real labs. This system promises to accelerate materials innovation by uniting literature, simulations, and lab feedback in one AI pipeline.
MIT researchers launched **CRESt**, a multimodal AI platform that integrates data from scientific texts, images, tables, and experiments. CRESt autonomously designs and runs lab experiments to discover new materials — for instance, predicting stable compounds or suggesting synthesis routes. It learns from rich multimodal sources and closes the loop by validating in real labs. This system promises to accelerate materials innovation by uniting literature, simulations, and lab feedback in one AI pipeline.