Kevin Murphy's Solo Paper: Using LLMs and Bayesian Methods for Efficient Scientific Discovery
burny_tech · x · 2026-08-14
Senior researcher Kevin Murphy published a solo-authored paper discussing how to discover the correct mechanistic model with minimal experiments when they are expensive. The approach introduces the Model Discovery Agent (MDA).
The core workflow includes:
- LLMs propose hypotheses: Generating potential explanations using large language models.
- Bayesian inference: Deciding what to believe about current hypotheses.
- Bayesian experimental design: Selecting the experiment that best separates competing explanations.
- M-open mechanism: Expanding the model space when all current hypotheses fail, rather than forcing a winner.
The author notes that in the AGI era, this approach to maximizing data efficiency is highly inspiring, wondering if this Bayesian instinct could eventually be distilled into the models themselves.
More from Research
- MLCommons launches MLPerf Endpoints, a benchmark that measures inference endpoints the way buyers actually purchase them — TheKanter · 2026-10-16
- DIY LEGO bench tests AI brick-building, GPT-6 Astra beats Opus — TheLegoless · 2026-10-16
- tszzl's theory: Tornado → Apache license → apaches → Southwest → Colorado — tszzl · 2026-10-16
- DIRECT: retrieval-based next-question suggestions for conversational agents — burkov · 2026-10-16
- A Qwen3 neuron tracks both Colorado and Tornado HTTP code; tszzl offers a theory — tszzl · 2026-10-16
- AI-Driven Timing-Aware Power Optimization Uses Learned Ranking With Baseline-Preserving Repair — Amit Sharma · 2026-10-16