Living in the Moment: Can Large Language Models Grasp Co-Temporal Reasoning?

TL;DR

This paper introduces CoTempQA, a new benchmark for evaluating large language models' ability to understand and reason about concurrent and intertwined events in time, revealing current limitations and proposing mathematical reasoning strategies.

Contribution

The paper presents CoTempQA, a comprehensive co-temporal reasoning benchmark, and explores mathematical reasoning methods to enhance LLMs' understanding of concurrent events.

Findings

LLMs perform significantly worse than humans on CoTempQA.

Chain of Thought methods do not fully bridge the reasoning gap.

Mathematical reasoning strategies can improve LLMs' co-temporal understanding.

Abstract

Temporal reasoning is fundamental for large language models (LLMs) to comprehend the world. Current temporal reasoning datasets are limited to questions about single or isolated events, falling short in mirroring the realistic temporal characteristics involving concurrent nature and intricate temporal interconnections. In this paper, we introduce CoTempQA, a comprehensive co-temporal Question Answering (QA) benchmark containing four co-temporal scenarios (Equal, Overlap, During, Mix) with 4,748 samples for evaluating the co-temporal comprehension and reasoning abilities of LLMs. Our extensive experiments reveal a significant gap between the performance of current LLMs and human-level reasoning on CoTempQA tasks. Even when enhanced with Chain of Thought (CoT) methodologies, models consistently struggle with our task. In our preliminary exploration, we discovered that mathematical reasoning plays a significant role in handling co-temporal events and proposed a strategy to boost LLMs' co-temporal reasoning from a mathematical perspective. We hope that our CoTempQA datasets will encourage further advancements in improving the co-temporal reasoning capabilities of LLMs. Our code is available at https://github.com/zhaochen0110/Cotempqa.