Monofact rate in training data governs LLM scientific discovery, preprint finds
CatAstro_Piyush · x · 2026-11-04
An arXiv preprint (2611.03617) proposes that "monofacts" — facts observed only once during training — set an LLM's capacity for scientific discovery. In materials science and regulatory genomics, raising monofact rates consistently yielded more discoveries across two open-weight model families, sampling strategies, and reasoning budgets. The effect isn't explained by training-set diversity alone and transfers across scientific domains when paired with in-domain data.
More from Research
- Anthropic says it may imminently hit RSP threshold for AI-accelerated R&D, releases provable inference work — tobyordoxford · 2026-11-04
- Google scanned 1,033 real household products, packed into a 345 MB GLB for three.js — victormustar · 2026-11-04
- These 10 GPU optimization papers are why your model runs fast today — NandoDF · 2026-11-04
- INT8 isn't the problem: I benchmarked samplers on INT8 flow models — YakovSava · 2026-11-04
- Karpathy Boosts Open Hacker News Dataset: 10M+ Posts Usable as a Personal Knowledge Base — huangyun_122 · 2026-11-04
- PixelCraft: A Multi-Agent Framework Targets Multimodal AI's Chart-Reading Failures — burkov · 2026-11-04