Year: 2023
Type: conference-paper
Abstract: Document retrieval is a key stage of standard Web search engines.Existing dual-encoder dense retrievers obtain representations for questions and documents independently, allowing for only shallow interactions between them.To overcome this limitation, recent autoregressive search engines replace the dual-encoder architecture by directly generating identifiers for relevant documents in the candidate pool.However, the training cost of such autoregressive search engines rises sharply as the number of candidate documents increases.In this paper, we find that large language models (LLMs) can follow human instructions to directly generate URLs for document retrieval.Surprisingly, when providing a few Query-URL pairs as in-context demonstrations, LLMs can generate Web URLs where nearly 90% of the corresponding documents contain correct answers to open-domain questions.In this way, LLMs can be thought of as built-in search engines, since they have not been explicitly trained to map questions to document identifiers.Experiments demonstrate that our method can consistently achieve better retrieval performance than existing retrieval approaches by a significant margin on three open-domain question answering benchmarks, under both zero and few-shot settings.The code for this work can be found at https://github.com/Ziems/ llm-url.
Authors: Noah Ziems, Wenhao Yu, Zhihan Zhang, Meng Jiang
Institution: University of Notre Dame
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
Awards: RI: Small: Learning to Retrieve Structured Information for Summarization and Translation of Unstructured Text, SHF: Medium: Collaborative Research: Semantically-Enhanced Software Traceability for Supporting Human-Centric Tasks, CAREER: Synergistic Approaches for Specialized Intelligent Assistance, G495881888, Collaborative Research: Advancing STEM Online Learning by Augmenting Accessibility with Explanatory Captions and AI
OpenAlex ID: W4385570369
