Can Long-Context Language Models Subsume Retrieval, RAG, SQL, and More?

TL;DR

Long-context language models can potentially replace traditional retrieval and reasoning systems by processing extensive information directly, as demonstrated by the LOFT benchmark showing competitive performance and highlighting areas for improvement.

Contribution

This paper introduces LOFT, a new benchmark for evaluating long-context language models on real-world tasks with extensive context, and demonstrates their surprising ability to rival specialized systems.

Findings

LCLMs can match state-of-the-art retrieval and RAG systems without explicit training.

Performance is heavily influenced by prompting strategies.

Challenges remain in compositional reasoning tasks like SQL.

Abstract

Long-context language models (LCLMs) have the potential to revolutionize our approach to tasks traditionally reliant on external tools like retrieval systems or databases. Leveraging LCLMs' ability to natively ingest and process entire corpora of information offers numerous advantages. It enhances user-friendliness by eliminating the need for specialized knowledge of tools, provides robust end-to-end modeling that minimizes cascading errors in complex pipelines, and allows for the application of sophisticated prompting techniques across the entire system. To assess this paradigm shift, we introduce LOFT, a benchmark of real-world tasks requiring context up to millions of tokens designed to evaluate LCLMs' performance on in-context retrieval and reasoning. Our findings reveal LCLMs' surprising ability to rival state-of-the-art retrieval and RAG systems, despite never having been explicitly trained for these tasks. However, LCLMs still face challenges in areas like compositional reasoning that are required in SQL-like tasks. Notably, prompting strategies significantly influence performance, emphasizing the need for continued research as context lengths grow. Overall, LOFT provides a rigorous testing ground for LCLMs, showcasing their potential to supplant existing paradigms and tackle novel tasks as model capabilities scale.