Are LLMs Prescient? A Continuous Evaluation using Daily News as the Oracle
This addresses the need for up-to-date and temporally-aware benchmarks for LLM evaluation, though it is incremental as it builds on existing prediction and RAG methods.
The authors tackled the problem of outdated LLM evaluation benchmarks by proposing a continuous evaluation method using daily news to assess temporal generalization and forecasting abilities, finding that LLM performance degrades over time as pre-training data becomes outdated, with RAG offering limited mitigation.
Many existing evaluation benchmarks for Large Language Models (LLMs) quickly become outdated due to the emergence of new models and training data.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 Our findings reveal that as pre-training data becomes outdated, LLM performance degrades over time.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.