HARPA: A Testability-Driven, Literature-Grounded Framework for Research Ideation
This addresses the problem of generating feasible and grounded research ideas for AI-driven scientific discovery, representing an incremental step forward in the field.
The paper tackled the challenge of generating testable and literature-grounded hypotheses for automated scientific discovery by developing HARPA, a framework that mines literature, explores design spaces, and converges on precise hypotheses. Results showed significant gains in feasibility (+0.78) and groundedness (+0.85) compared to a baseline, with HARPA producing more successful executions (20 vs. 11 out of 40) and learning from prior outcomes to improve hypothesis scoring by approximately 28%.
While there has been a surge of interest in automated scientific discovery (ASD), especially with the emergence of LLMs, it remains challenging for tools to generate hypotheses that are both testable and grounded in the scientific literature. Additionally, existing ideation tools are not adaptive to prior experimental outcomes. We developed HARPA to address these challenges by incorporating the ideation workflow inspired by human researchers. HARPA first identifies emerging research trends through literature mining, then explores hypothesis design spaces, and finally converges on precise, testable hypotheses by pinpointing research gaps and justifying design choices. Our evaluations show that HARPA-generated hypothesis-driven research proposals perform comparably to a strong baseline AI-researcher across most qualitative dimensions (e.g., specificity, novelty, overall quality), but achieve significant gains in feasibility(+0.78, p$<0.05$, bootstrap) and groundedness (+0.85, p$<0.01$, bootstrap) on a 10-point Likert scale. When tested with the ASD agent (CodeScientist), HARPA produced more successful executions (20 vs. 11 out of 40) and fewer failures (16 vs. 21 out of 40), showing that expert feasibility judgments track with actual execution success. Furthermore, to simulate how researchers continuously refine their understanding of what hypotheses are both testable and potentially interesting from experience, HARPA learns a reward model that scores new hypotheses based on prior experimental outcomes, achieving approx. a 28\% absolute gain over HARPA's untrained baseline scorer. Together, these methods represent a step forward in the field of AI-driven scientific discovery.