Notes2Skills: From Lab Notebooks to Certainty-Aware Scientific Agent Skills

2026-06-10

AI Summary

This work addresses the challenge that existing AI methods struggle to distinguish between definitive observations, uncertain judgments, and implicit procedural steps in laboratory notebooks, often misinterpreting tentative content as certain instructions. To overcome this limitation, the authors propose the first certainty-aware, two-stage framework tailored for scientific AI agents. The approach first employs natural language processing to parse unstructured notes and then integrates certainty modeling with a validation pipeline to produce structured skill instructions annotated with confidence levels. Evaluated across seven experimental conditions and three wet-lab trials, this method is the only one that simultaneously avoids misusing uncertain information while preserving all definitive content, thereby significantly enhancing the reliability and safety of AI-augmented collaborative research systems.

Abstract

Scientific discovery workflows usually contain and rely heavily on lab notes, where researchers record observations, interpret uncertain results, and plan follow-up experiments. Such informative lab notes preserve evolving scientific reasoning and author uncertainty, rather than polished final results exhibited in publications, providing a valuable opportunity for AI to engage in scientific exploration at a more comprehensive and deeper level. However, most prior work on scientific text focuses on papers, protocols, or structured databases, leaving informal laboratory notes underexplored as inputs to AI agents for science. This gap matters because lab notes often intermingle validated observations, tentative judgments, and possible experimental next steps within the same passage. If these signals are conflated, an AI agent may mistake uncertain scientific judgments for confirmed conclusions or executable actions. To this end, we present Notes2Skills, a two-stage framework for turning lab notebooks into verifiable skills for scientific AI agents while preserving the author's certainty. Across seven conditions and three wet-lab sessions, Notes2Skills is the only configuration that neither mistakes uncertain notes for firm instructions nor discards firm ones. We show that certainty preservation is the missing piece between lab notebooks and reliable agent skills, opening a path toward safer AI co-scientist systems.

Problem

• lab notebooks
• scientific AI agents
• uncertainty awareness
• scientific reasoning
• agent skills

Innovation

• lab notebooks
• scientific AI agents
• certainty-aware
• skill extraction
• uncertainty preservation