Open-ended Scientific Discovery with Possibilistic Reasoning

2026-10-08

AI Summary

This study addresses the challenge of balancing hypothesis iteration with statistical validity in autonomous scientific discovery using large language models (LLMs). To this end, it proposes a possibility theory-based framework for Anytime Autonomous Scientific Discovery (AASD), providing its first formal definition. The framework introduces an abductive utility metric and a possibility frontier search algorithm, which, combined with LLMs, enable adaptive hypothesis generation. This approach ensures anytime validity while asymptotically achieving optimality. Experimental results demonstrate that the proposed method performs exceptionally well on both synthetic and real-world scientific tasks, effectively advancing the progress of autonomous scientific discovery.

Abstract

Autonomous scientific discovery with LLMs requires generating and testing hypotheses adaptively as evidence accumulates while maintaining statistical validity. Existing anytime-valid methods can handle data-dependent hypotheses, but open-ended discovery poses a deeper challenge: the best discovered hypothesis may still be the best of a bad lot, with better explanations yet undiscovered, while even background knowledge such as physical laws may require revision in light of new findings. In response, we formalize the problem as Abductive Autonomous Scientific Discovery (AASD) using possibility theory. We introduce abductive utility, a computable measure of discovery progress, and possibility frontier search, the first algorithm for AASD, which maintains anytime validity and achieves $\varepsilon$-optimal abductive utility asymptotically under suitable conditions. Experiments on synthetic and real-world scientific-discovery tasks show strong performance.

Problem

• Autonomous Scientific Discovery
• Open-ended Discovery
• Abductive Reasoning
• Anytime Validity
• Possibility Theory

Innovation

• Abductive Autonomous Scientific Discovery
• Possibility Theory
• Abductive Utility
• Possibility Frontier Search
• Anytime Validity