HypER: Literature-grounded Hypothesis Generation and Distillation with Provenance

TL;DR

HypER is a specialized small language model designed for literature-grounded hypothesis generation, emphasizing reasoning and evidence provenance, outperforming baseline models in validity discrimination and hypothesis quality.

Contribution

It introduces HypER, a multi-task trained model that improves scientific hypothesis generation by incorporating reasoning and evidence provenance, surpassing existing methods.

Findings

HypER outperforms base models in reasoning accuracy (+22% F1).

Generates more evidence-grounded hypotheses with higher quality scores.

Achieves high feasibility and impact ratings from human experts.

Abstract

Large Language models have demonstrated promising performance in research ideation across scientific domains. Hypothesis development, the process of generating a highly specific declarative statement connecting a research idea with empirical validation, has received relatively less attention. Existing approaches trivially deploy retrieval augmentation and focus only on the quality of the final output ignoring the underlying reasoning process behind ideation. We present $\texttt{HypER}$ ($\textbf{Hyp}$othesis Generation with $\textbf{E}$xplanation and $\textbf{R}$easoning), a small language model (SLM) trained for literature-guided reasoning and evidence-based hypothesis generation. $\texttt{HypER}$ is trained in a multi-task setting to discriminate between valid and invalid scientific reasoning chains in presence of controlled distractions. We find that $\texttt{HypER}$ outperformes the base model, distinguishing valid from invalid reasoning chains (+22\% average absolute F1), generates better evidence-grounded hypotheses (0.327 vs. 0.305 base model) with high feasibility and impact as judged by human experts ($>$3.5 on 5-point Likert scale).