SCITAT: A Question Answering Benchmark for Scientific Tables and Text Covering Diverse Reasoning Types

TL;DR

SciTaT is a comprehensive benchmark for scientific question answering that includes diverse reasoning types and emphasizes the integration of tables and text, aiming to better reflect real-world scientific inquiry.

Contribution

The paper introduces SciTaT, a new benchmark dataset with diverse reasoning types and combined table-text questions, along with a strong baseline model CaR that improves reasoning performance.

Findings

CaR achieves 12.9% higher accuracy than existing baselines.

SciTaT covers more reasoning types and better reflects real scientific questions.

Challenges include complex calculations and domain-specific knowledge.

Abstract

Scientific question answering (SQA) is an important task aimed at answering questions based on papers. However, current SQA datasets have limited reasoning types and neglect the relevance between tables and text, creating a significant gap with real scenarios. To address these challenges, we propose a QA benchmark for scientific tables and text with diverse reasoning types (SciTaT). To cover more reasoning types, we summarize various reasoning types from real-world questions. To involve both tables and text, we require the questions to incorporate tables and text as much as possible. Based on SciTaT, we propose a strong baseline (CaR), which combines various reasoning methods to address different reasoning types and process tables and text at the same time. CaR brings average improvements of 12.9% over other baselines on SciTaT, validating its effectiveness. Error analysis reveals the challenges of SciTaT, such as complex numerical calculations and domain knowledge.