Promptriever: Instruction-Trained Retrievers Can Be Prompted Like Language Models

TL;DR

Promptriever is a novel retrieval model trained with instruction data that can be prompted like language models, achieving state-of-the-art results and demonstrating controllability and robustness in retrieval tasks.

Contribution

This work introduces Promptriever, the first retrieval model that can be prompted like an LM, trained on a new instruction dataset from MS MARCO, and capable of following detailed instructions.

Findings

Achieves state-of-the-art on FollowIR with +14.3 p-MRR

Increases robustness to query phrasing by +12.9 Robustness@10

Enables hyperparameter tuning via prompting, improving BEIR scores by +1.4

Abstract

Instruction-tuned language models (LM) are able to respond to imperative commands, providing a more natural user interface compared to their base counterparts. In this work, we present Promptriever, the first retrieval model able to be prompted like an LM. To train Promptriever, we curate and release a new instance-level instruction training set from MS MARCO, spanning nearly 500k instances. Promptriever not only achieves strong performance on standard retrieval tasks, but also follows instructions. We observe: (1) large gains (reaching SoTA) on following detailed relevance instructions (+14.3 p-MRR / +3.1 nDCG on FollowIR), (2) significantly increased robustness to lexical choices/phrasing in the query+instruction (+12.9 Robustness@10 on InstructIR), and (3) the ability to perform hyperparameter search via prompting to reliably improve retrieval performance (+1.4 average increase on BEIR). Promptriever demonstrates that retrieval models can be controlled with prompts on a per-query basis, setting the stage for future work aligning LM prompting techniques with information retrieval.