A Brief Survey on Temporal Reasoning Based on Large Language Models
Panfeng Zhang, Huan Zhang, Xiaoke Wang, Fu Zhang, Fan Yu · 2024
Temporal reasoning is a pivotal mechanism for understanding the world around us, enabling inference, prediction, and deduction of temporal relationships among events. The advent of Large Language Models (LLMs) has sparked considerable interest in research on temporal reasoning utilizing these models. These models, trained on massive datasets, acquire potent representational capabilities, allowing them to learn temporal patterns and perform inference and prediction on complex temporal data. We provide a brief overview of recent research on temporal reasoning based on LLMs, exploring the capabilities of LLMs in temporal reasoning and outlining future directions. We particularly focus on four major research areas: Time Series Forecasting, Temporal Question Answering, Temporal Knowledge Graph and Assessing Temporal Reasoning Capability in LLMs. Through this review, we aim to offer new insights and perspectives for research and applications in the field of temporal reasoning, further advancing research on LLMs in temporal reasoning.