Analyzing Temporal Complex Events with Large Language Models? A Benchmark towards Temporal, Long Context Understanding

TL;DR

This paper introduces TCELongBench, a benchmark for evaluating large language models on their ability to understand and analyze temporal complex events over long contexts, focusing on event extraction, sequencing, and forecasting.

Contribution

It presents a new benchmark and evaluation framework for LLMs to handle temporal complex events, including tasks and a retrieval-augmented approach for long texts.

Findings

Models with effective retrievers perform comparably to those with long context windows.

The benchmark covers reading comprehension, temporal sequencing, and future event prediction.

Long context windows improve performance on complex event understanding.

Abstract

The digital landscape is rapidly evolving with an ever-increasing volume of online news, emphasizing the need for swift and precise analysis of complex events. We refer to the complex events composed of many news articles over an extended period as Temporal Complex Event (TCE). This paper proposes a novel approach using Large Language Models (LLMs) to systematically extract and analyze the event chain within TCE, characterized by their key points and timestamps. We establish a benchmark, named TCELongBench, to evaluate the proficiency of LLMs in handling temporal dynamics and understanding extensive text. This benchmark encompasses three distinct tasks - reading comprehension, temporal sequencing, and future event forecasting. In the experiment, we leverage retrieval-augmented generation (RAG) method and LLMs with long context window to deal with lengthy news articles of TCE. Our findings indicate that models with suitable retrievers exhibit comparable performance with those utilizing long context window.