Bio papers never deliver what abstracts claim — a worse substrate for LLMs
anshulkundaje · x · 2026-09-26
Follow-up in the discussion: hallucination is especially bad in biology because no study ever truly delivers what its abstract claims — such studies simply don't work in scenarios with multiple conflicting interpretations. When a field's literature carries systematic bias, LLMs trained on or retrieving from it amplify the problem.
More from Models
- Perceptron launches Mk1.5, one embodied AI model for drones, quadrupeds and smart glasses — code_star · 2026-09-26
- Tester: Astra is "autistic" at parsing human sentiment; Fable and Opus run circles around it — teortaxesTex · 2026-09-26
- User generates an anime-style fight scene entirely with code using Claude Opus 5.5 — EricBuess · 2026-09-26
- Dev verdict on Opus 5.5: the first good Opus since 4.8 — holdenmatt · 2026-09-26
- Opus 5.5 code review demo draws attention — tlakomy · 2026-09-26
- One prompt, zero libraries: Opus 5.5 builds polished vanilla JS motion graphics in Claude Code — EricBuess · 2026-09-26