Overview

Scientific discovery studies how to turn observations and prior knowledge into new, testable explanations or useful candidates, from physical laws and biological mechanisms to materials and research questions. Common approaches include symbolic regression and causal or mechanistic modeling, machine-learning analysis, Bayesian optimization for choosing experiments, and language-model agents that search literature, write code, or coordinate tests. The field is moving from passive pattern fitting toward systems that propose hypotheses, design informative measurements, revise models when evidence disagrees, and verify results. A central challenge is that accurate predictions or plausible equations do not guarantee recovered mechanisms or genuinely new insight, so evaluations increasingly test robustness, evidence grounding, experimental efficiency, and uncertainty.

Latest papers

91 papers9 papers in the last four weeks, down 47% on the four weeks before. 0.1% of all new papers.

Jun 20, 2025

Discrete Compositional Generation via General Soft Operators and Robust Reinforcement Learning

A major bottleneck in scientific discovery consists of narrowing an exponentially large set of objects, such as proteins or molecules, to a small set of promising candidates with desirable properties. While this process can rely on expert knowledge, recent methods leverage reinforcement learning (RL) guided by a proxy reward function to enable this filtering. By employing various forms of entropy regularization, these methods aim to learn samplers that generate diverse candidates that are highly rated by the proxy function. In this work, we make two main contributions. First, we show that these methods are liable to generate overly diverse, suboptimal candidates in large search spaces. To address this issue, we introduce a novel unified operator that combines several regularized RL operators into a general framework that better targets peakier sampling distributions. Secondly, we offer a novel, robust RL perspective of this filtering process. The regularization can be interpreted as robustness to a compositional form of uncertainty in the proxy function (i.e., the true evaluation of a candidate differs from the proxy's evaluation). Our analysis leads us to a novel, easy-to-use algorithm we name trajectory general mellowmax (TGM): we show it identifies higher quality, diverse candidates than baselines in both synthetic and real-world tasks. Code: https://github.com/marcojira/tgm.