Go to ICLR 2025 Conference homepageRARe: Retrieval Augmented Retrieval with In-Context Examples
Keywords: Retrieval, Embedding models, In-Context Learning
Abstract: We investigate whether in-context examples, widely used in decoder-only language models (LLMs), can improve embedding models for retrieval. Unlike in LLMs, naively prepending in-context examples (query-document pairs) to the target query at inference time does not work out of the box. We introduce a simple approach to enable retrievers to use in-context examples. Our approach, \texttt{RARe}, fine-tunes a pre-trained model with in-context examples whose query is semantically similar to the target query. This can be applied to adapt various base architectures (i.e., decoder-only language models, retriever models) and consistently achieves performance gains of up to +2.72\% nDCG across various open-domain retrieval datasets (BeIR, RAR-b). Particularly, we find \texttt{RARe} exhibits stronger out-of-domain generalization compared to models using queries without in-context examples, similar to what is seen for in-context learning in LLMs. While our approach incurs additional computational cost to encode lengthier queries, the impact is less pronounced in large-corpus scenarios. We further provide analysis on the design choices of in-context example augmentation and lay the foundation for future work in this space.
Primary Area: applications to computer vision, audio, language, and other modalities
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 10674
Loading