1/ Updates on our improved open-source CheXagent with new transparent benchmarks! We ran a new reader study mimicking real workflows: Radiology residents drafted reports that attendings reviewed/edited. Results from 8 rads show major efficiency gains. Key findings: 👇

💥 Gemini 2.0 is on paper-central. Talk with any paper from the 🤗 Hugging Face paper page. Example with GenEx 👇

2025 will be the year of AI for science Leveraging all the things we've recently learned training AI models for 1000x impact in science and this will need data! We're excited to team up with the team at @entalpic_ai for LeMaterial -- an open source initiative aiming to facilitate (AI for) materials discovery releasing datasets, hash function and tools to explore the chemical space for material design In this first release: - data: releasing LeMat-Bulk, a 6.7M dataset of solid-state materials which cleans, standardizes and unifies Materials Project, Alexandria and OQMD - exploration tools: comprehensive suit to analyze the chemical space, including phase diagrams, visualization tools or elemental composition treemap, empowering researchers to analyze and understand materials at scale - benchmarks tools: hashing function creating material fingerprints --designed to deduplicate data and evaluate the "novelty" of newly generated materials. It attempts to answer the question « what is a material », serving as basis for discussion