BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Mapping Materials Science: a multi-modal toolbox to curate broad s
 ynthesis procedure databases from scientific literature
DTSTART:20260428T151500
DTEND:20260428T161500
DTSTAMP:20260506T173153Z
UID:6d0cccb564c66cb53c17a7fc486ad8f8e5d27c83904d92255a798b6e
CATEGORIES:Conferences - Seminars
DESCRIPTION:Magdalena Lederbauer\nAbstract\nPredicting how to synthesize a
  material is a fundamental challenge in materials discovery\, since proced
 ural knowledge is scattered across decades of literature in formats inacce
 ssible to data-driven methods. Without linking how a material is made to h
 ow well it performs\, predictive models cannot learn to optimize synthesis
  procedures for target properties.\nHere\, we present LeMat-Synth\, a modu
 lar open-source toolbox that transforms unstructured materials science lit
 erature into linked\, machine-readable synthesis-performance databases at 
 scale. Applied to 81k open-access papers\, LeMat-Synth curates structured 
 synthesis procedures spanning 35 synthesis methods and 16 material classes
 . We demonstrate generalizability through two case studies in thermocataly
 sis for ammonia decomposition and superconductor discovery\, where linked 
 synthesis protocols and digitized performance figures enable data-driven i
 nsights previously invisible at scale. Together\, these position LeMat-Syn
 th as a generalizable data infrastructure layer that aims to contribute to
  autonomous materials discovery.\n\nBiography\nMagdalena Lederbauer is a P
 hD student in chemical engineering and computer science at MIT\, working a
 t the intersection of machine learning and the chemical sciences. Her rese
 arch spans data infrastructure for materials discovery\, structure elucida
 tion from mass spectrometry and reaction discovery in catalysis. Trained a
 s a chemist at ETH Zurich\, she received the ETH Medal\, S.&N. Blank Prize
  and the Willi Studer Prize for her MSc. As an Entalpic Research Fellow\, 
 she leads the working group for large-language-model-driven synthesis data
  extraction at LeMaterial\, an open-source initiative in collaboration wit
 h Hugging Face that builds a community-driven ecosystem for materials scie
 nce data and artificial intelligence.
LOCATION:https://epfl.zoom.us/j/68447908297?pwd=OU5JUGJUSUhZc0ZNYjQ2WENvYI
 NRdz09
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR