Enhancing Study-Level Inference from Clinical Trial Papers via Reinforcement Learning-Based Numeric Reasoning

TL;DR

This paper introduces a reinforcement learning-based numeric reasoning system to improve the extraction of quantitative evidence from clinical trial papers, leading to more accurate and interpretable study-level conclusions in systematic reviews.

Contribution

It presents a novel reinforcement learning approach for numeric data extraction and reasoning, outperforming retrieval-based systems and large language models on relevant benchmarks.

Findings

Up to 21% absolute F1 score improvement over retrieval systems

Outperforms large language models by up to 9% on RCTs benchmark

Reinforcement learning enhances numeric reasoning accuracy

Abstract

Systematic reviews in medicine play a critical role in evidence-based decision-making by aggregating findings from multiple studies. A central bottleneck in automating this process is extracting numeric evidence and determining study-level conclusions for specific outcomes and comparisons. Prior work has framed this problem as a textual inference task by retrieving relevant content fragments and inferring conclusions from them. However, such approaches often rely on shallow textual cues and fail to capture the underlying numeric reasoning behind expert assessments. In this work, we conceptualise the problem as one of quantitative reasoning. Rather than inferring conclusions from surface text, we extract structured numerical evidence (e.g., event counts or standard deviations) and apply domain knowledge informed logic to derive outcome-specific conclusions. We develop a numeric reasoning system composed of a numeric data extraction model and an effect estimate component, enabling more accurate and interpretable inference aligned with the domain expert principles. We train the numeric data extraction model using different strategies, including supervised fine-tuning (SFT) and reinforcement learning (RL) with a new value reward model. When evaluated on the CochraneForest benchmark, our best-performing approach -- using RL to train a small-scale number extraction model -- yields up to a 21% absolute improvement in F1 score over retrieval-based systems and outperforms general-purpose LLMs of over 400B parameters by up to 9% on the RCTs benchmark. Our results demonstrate the promise of reasoning-driven approaches for automating systematic evidence synthesis.