Search papers, labs, and topics across Lattice.
This paper introduces StatefulDiscovery, a framework designed to enhance open-ended scientific discovery by addressing the evidence-calibration problem, where agents must balance exploration with the validity of claims made. By externalizing investigation state, the framework effectively coordinates the selection of research frontiers, evidence gathering, and the adjudication of claims across 40 real-data discovery tasks. The results show that StatefulDiscovery generates a higher number of claims that are both well-supported and high-value compared to several baseline methods, highlighting the importance of structured hypotheses and local adjudication in the discovery process.
StatefulDiscovery significantly boosts the generation of high-value scientific claims by effectively coupling exploration with evidence calibration.
Open-ended scientific discovery asks agents to move beyond executing analyses for predefined questions. Across multiple rounds of exploration, a discovery agent must decide which phenomena warrant investigation while avoiding overinterpretation, where emerging claims exceed the evidential scope of the analyses supporting them. This creates an evidence-calibration problem: the exploration trajectory must be coupled with claim status so that evidence can guide both what to investigate next and what can be claimed. We introduce StatefulDiscovery, a discovery framework that externalizes investigation state and uses it to coordinate frontier selection, evidence acquisition, and claim adjudication. We evaluate StatefulDiscovery across 40 real-data discovery tasks. Compared with several baselines, StatefulDiscovery produces more claims overall judged to be both well-supported and high-value. Ablations indicate that structured hypotheses, local adjudication, and frontier control contribute to performance. Together, these results suggest that explicit discovery state can couple exploration with evidence-calibrated claim formation.
