Are LLMs Prescient? A Continuous Evaluation using Daily News as the Oracle

TL;DR

This paper introduces Daily Oracle, a continuous evaluation benchmark using daily news to assess LLMs' ability to predict future events, revealing performance degradation over time and the potential of retrieval augmentation.

Contribution

It proposes a novel, dynamic benchmark for evaluating LLMs' temporal prediction abilities using news data, addressing limitations of static benchmarks.

Findings

LLM performance declines as pre-training data becomes outdated

Retrieval Augmented Generation improves prediction accuracy

Performance degradation persists despite retrieval augmentation

Abstract

Many existing evaluation benchmarks for Large Language Models (LLMs) quickly become outdated due to the emergence of new models and training data. These benchmarks also fall short in assessing how LLM performance changes over time, as they consist of a static set of questions without a temporal dimension. To address these limitations, we propose using future event prediction as a continuous evaluation method to assess LLMs' temporal generalization and forecasting abilities. Our benchmark, Daily Oracle, automatically generates question-answer (QA) pairs from daily news, challenging LLMs to predict "future" event outcomes. Our findings reveal that as pre-training data becomes outdated, LLM performance degrades over time. While Retrieval Augmented Generation (RAG) has the potential to enhance prediction accuracy, the performance degradation pattern persists, highlighting the need for continuous model updates. Code and data are available at https://agenticlearning.ai/daily-oracle.