Lab is pre-training clinical models, but outsiders assume it's just fine-tuning
madeofAjala · x · 2026-10-24
Zeeskylaw shares that their lab is pre-training clinical-domain models, yet nearly everyone outside assumes they're just fine-tuning an existing clinical NLP model. "Communication is really half of research," the author notes.
More from Research
- Nvidia and UW Institute for Protein Design open-source RFD4-Proteina protein design model — AllThingsApx · 2026-10-24
- Ben Goertzel argues the selfish-supermind attractor has a wider basin than BGI — bengoertzel · 2026-10-24
- 11 top mathematicians plus Google's Astra solve major 2+1D KPZ equation problem — ChrSzegedy · 2026-10-24
- Diffusion LMs allocate compute by task: novel outputs cost more forward passes — uclanlp · 2026-10-24
- New paper: LLM simulators can evaluate AI control protocols without real deployments — jammastergirish · 2026-10-24
- FrontierCS 2 benchmark makes agents beat original methods from code with the paper's contribution deleted — berkeley_ai · 2026-10-24