Cristina Sarasua

Author directory

2026

Recent advances in large language models (LLMs) have accelerated automated scientific discovery (ASD), but existing ideation tools still struggle to generate hypotheses that are both testable, literature-grounded, and adaptive to prior experimental outcomes. We developed HARPA to address these challenges by following a human-inspired ideation workflow that mines literature trends, explores hypothesis design spaces, and converges on precise, testable hypotheses by pinpointing research gaps and justifying design choices. Human evaluations show that HARPA-generated hypothesis-driven research proposals performs comparably to a strong baseline system, AI-researcher, in novelty and overall quality, while showing a positive trend in feasibility (+0.78, p<0.05, bootstrap) and a significant gain in groundedness (+0.85, p<0.01, bootstrap) on a 10-point Likert scale. When evaluated 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) indicating that expert feasibility judgments track execution success. HARPA also learns a reward model that scores hypotheses based on prior experimental outcomes, achieving approx. a 28% absolute gain over HARPA’s untrained baseline scorer.

2025

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 HypER (Hypothesis Generation with Explanation and Reasoning), a small language model (SLM) trained for literature-guided reasoning and evidence-based hypothesis generation. 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 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).

2023

Current approaches to Argument Mining (AM) tend to take a holistic or black-box view of the overall pipeline. This paper, in contrast, aims to provide a solution to achieve increased performance based on current components instead of independent all-new solutions. To that end, it presents the Deployment of Recombination and Ensemble methods for Argument Miners (DREAM) framework that allows for the (automated) combination of AM components. Using ensemble methods, DREAM combines sets of AM systems to improve accuracy for the four tasks in the AM pipeline. Furthermore, it leverages recombination by using different argument miners elements throughout the pipeline. Experiments with five systems previously included in a benchmark show that the systems combined with DREAM can outperform the previous best single systems in terms of accuracy measured by an AM benchmark.