TimeBench: A Comprehensive Evaluation of Temporal Reasoning Abilities in Large Language Models

TL;DR

TimeBench is a comprehensive benchmark designed to evaluate the temporal reasoning abilities of large language models across various phenomena, revealing significant gaps compared to human performance and highlighting challenges in the field.

Contribution

This paper introduces TimeBench, the first extensive hierarchical benchmark for temporal reasoning in large language models, covering diverse reasoning categories and providing a platform for future research.

Findings

LLMs lag behind humans in temporal reasoning tasks.

Performance varies across different reasoning categories.

Multiple factors influence LLMs' temporal reasoning capabilities.

Abstract

Grasping the concept of time is a fundamental facet of human cognition, indispensable for truly comprehending the intricacies of the world. Previous studies typically focus on specific aspects of time, lacking a comprehensive temporal reasoning benchmark. To address this, we propose TimeBench, a comprehensive hierarchical temporal reasoning benchmark that covers a broad spectrum of temporal reasoning phenomena. TimeBench provides a thorough evaluation for investigating the temporal reasoning capabilities of large language models. We conduct extensive experiments on GPT-4, LLaMA2, and other popular LLMs under various settings. Our experimental results indicate a significant performance gap between the state-of-the-art LLMs and humans, highlighting that there is still a considerable distance to cover in temporal reasoning. Besides, LLMs exhibit capability discrepancies across different reasoning categories. Furthermore, we thoroughly analyze the impact of multiple aspects on temporal reasoning and emphasize the associated challenges. We aspire for TimeBench to serve as a comprehensive benchmark, fostering research in temporal reasoning. Resources are available at: https://github.com/zchuz/TimeBench