Large Search Model: Redefining Search Stack in the Era of LLMs

TL;DR

This paper proposes a unified search framework using large language models that reformulates all search tasks as text generation problems, aiming to improve search quality and simplify the search architecture.

Contribution

It introduces the large search model framework that unifies search components with LLMs and formulates tasks as autoregressive text generation, a novel approach in search technology.

Findings

Proof-of-concept experiments demonstrate feasibility.

Framework leverages LLMs' reasoning capabilities.

Potential to simplify search system architecture.

Abstract

Modern search engines are built on a stack of different components, including query understanding, retrieval, multi-stage ranking, and question answering, among others. These components are often optimized and deployed independently. In this paper, we introduce a novel conceptual framework called large search model, which redefines the conventional search stack by unifying search tasks with one large language model (LLM). All tasks are formulated as autoregressive text generation problems, allowing for the customization of tasks through the use of natural language prompts. This proposed framework capitalizes on the strong language understanding and reasoning capabilities of LLMs, offering the potential to enhance search result quality while simultaneously simplifying the existing cumbersome search stack. To substantiate the feasibility of this framework, we present a series of proof-of-concept experiments and discuss the potential challenges associated with implementing this approach within real-world search systems.