BEGIN:VCALENDAR
PRODID:-//github.com/ical-org/ical.net//NONSGML ical.net 4.0//EN
VERSION:2.0
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:STANDARD
DTSTART:20241103T020000
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:EST
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20250309T020000
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:EDT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
DESCRIPTION:This session features two talks exploring novel approaches to 
 AI-driven biomedical discovery and drug representation. "PredMed: Extrapo
 lating Future Discoveries from the Literature Universe" by Chia-Hsuan Cha
 ng\, PhD - Postdoctoral Associate in Biomedical Informatics and Data Scie
 nce Abstract: Most AI co-scientists are limited by retrieval-augmented ge
 neration (RAG) over static corpora and heavily rely on human guidance. We
  present PredMed\, a novel framework that redefines hypothesis generation
  as a temporal extrapolation task within the high-dimensional literature 
 universe. Using time-based regression and a specialized Embedding Languag
 e Model (ELM) acting as a decoder\, we project and translate future-state
  embeddings back into natural language. Our results show that this tempor
 al steering mechanism explores scientific territory that standard prompti
 ng cannot reach\, outperforming baseline methods in both novelty and rela
 tional depth. We also validate PredMed’s efficacy through expert-reviewed
  hypotheses in CAR-T therapy domain\, highlighting a new frontier for aut
 onomous scientific discovery "A Literature-Based Drug Embedding Resource 
 for Biomedical Research" by Zhiyuan Cao - PhD Student in Computational Bi
 ology and Biomedical Informatics (CBB) Abstract: We introduce DrugSpace\,
  a reusable text-based drug embedding resource designed to support simila
 rity search\, retrieval\, and downstream modeling in biomedical research.
  Built from large-scale PubMed abstracts and aligned with heterogeneous D
 rugBank drug descriptions through a two-stage training pipeline\, DrugSpa
 ce is released both as a versioned embedding dataset and as an embedder f
 or generating representations from new drug text. To support realistic re
 use\, the resource is evaluated under a prospective setting that separate
 s drug-level alignment from later drug introductions and updates. Across 
 intrinsic similarity discrimination\, ATC-based therapeutic retrieval\, r
 obustness to input perturbations\, and integration into a representative 
 DDI prediction pipeline\, DrugSpace consistently remains competitive with
  strong biomedical and general-purpose text embedding baselines\, support
 ing its utility as a practical and extensible drug representation resourc
 e.\n\nSpeakers:\nChia-Hsuan Chang\; Zhiyuan Cao\n\nAdmission:\nFree\n\nDe
 tails URL:\nhttps://medicine.yale.edu/event/nlpllm-interest-group-28/\n
DTEND;TZID=America/New_York:20260323T170000
DTSTAMP:20260429T181345Z
DTSTART;TZID=America/New_York:20260323T160000
LOCATION:Subscribe to receive Zoom Passcode: https://mailman.yale.edu/mail
 man/listinfo/nlp-llm-ig \, URL: https://yale.zoom.us/j/93599941969
SEQUENCE:0
STATUS:Confirmed
SUMMARY:NLP/LLM Interest Group
UID:eddb53a3-6a87-4d3f-9061-e62ad3ff5873
END:VEVENT
END:VCALENDAR