ScholarChemQA: Unveiling the Power of Language Models in Chemical Research Question Answering

TL;DR

ScholarChemQA introduces a large-scale chemical question answering dataset and a specialized model that leverages data augmentation, re-weighting, and calibration to improve reasoning and understanding in chemical research questions.

Contribution

The paper presents ScholarChemQA, a novel chemical QA dataset, and QAMatch, a tailored model that addresses data imbalance and unlabeled data challenges in chemical question answering.

Findings

QAMatch outperforms recent baselines and LLMs on ScholarChemQA.

The dataset reflects real-world chemical research challenges.

Data augmentation and re-weighting improve model performance.

Abstract

Question Answering (QA) effectively evaluates language models' reasoning and knowledge depth. While QA datasets are plentiful in areas like general domain and biomedicine, academic chemistry is less explored. Chemical QA plays a crucial role in both education and research by effectively translating complex chemical information into readily understandable format. Addressing this gap, we introduce ScholarChemQA, a large-scale QA dataset constructed from chemical papers. This dataset reflects typical real-world challenges, including an imbalanced data distribution and a substantial amount of unlabeled data that can be potentially useful. Correspondingly, we introduce a QAMatch model, specifically designed to effectively answer chemical questions by fully leveraging our collected data. We first address the issue of imbalanced label distribution by re-weighting the instance-wise loss based on the inverse frequency of each class, ensuring minority classes are not dominated by majority ones during optimization. Next, we utilize the unlabeled data to enrich the learning process, generating a variety of augmentations based on a SoftMix operation and ensuring their predictions align with the same target, i.e., pseudo-labels. To ensure the quality of the pseudo-labels, we propose a calibration procedure aimed at closely aligning the pseudo-label estimates of individual samples with a desired ground truth distribution. Experiments show that our QAMatch significantly outperforms the recent similar-scale baselines and Large Language Models (LLMs) not only on our ScholarChemQA dataset but also on four benchmark datasets. We hope our benchmark and model can facilitate and promote more research on chemical QA.