EMNLP 2026 Study: LLMs Rely on Morphological Shortcuts for Drug Reasoning, Posing Safety Risks
jessyjli · x · 2026-08-22
Accepted to EMNLP 2026, this study finds LLMs often lack drug-specific knowledge and rely on word morphology. Models can confidently reason about fictitious drugs via affixes, a behavior localized to early-mid layers that poses safety risks.
More from Research
- Anthropic Mathematician Claims Claude Solved Open Math Problem from 1948 — GregCook2011 · 2026-08-24
- New Architecture RHEA: Train 1B Model on 8GB VRAM — zemondza · 2026-08-24
- Trained two 16M-param models to do generative CAD with real physics — debreuil · 2026-08-24
- Claude model helps discover complex structure on S^6, solving 60-year-old math problem — Singularitarian · 2026-08-24
- Study: Agents read instructions/notes 60.5% of the time, rarely touch API docs — dair_ai · 2026-08-24
- Claude Verifies 43 Lean Modules autonomously, Tackling Theoretical Physics — Tkaraletsos · 2026-08-24