AGRaME: Any-Granularity Ranking with Multi-Vector Embeddings

Revanth Gangi Reddy, Omar Attia, Yunyao Li, Heng Ji, Saloni Potdar


Abstract
Ranking is a fundamental problem in search, however, existing ranking algorithms usually restrict the granularity of ranking to full passages or require a specific dense index for each desired level of granularity. Such lack of flexibility in granularity negatively affects many applications that can benefit from more granular ranking, such as sentence-level ranking for open-domain QA, or proposition-level ranking for attribution. In this work, we introduce the idea of any-granularity ranking which leverages multi-vector embeddings to rank at varying levels of granularity while maintaining encoding at a single (coarser) level of granularity. We propose a multi-granular contrastive loss for training multi-vector approaches and validate its utility with both sentences and propositions as ranking units. Finally, we demonstrate the application of proposition-level ranking to post-hoc citation addition in retrieval-augmented generation, surpassing the performance of prompt-driven citation generation.
Anthology ID:
2024.emnlp-main.490
Volume:
Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing
Month:
November
Year:
2024
Address:
Miami, Florida, USA
Editors:
Yaser Al-Onaizan, Mohit Bansal, Yun-Nung Chen
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
8630–8641
Language:
URL:
https://preview.aclanthology.org/build-pipeline-with-new-library/2024.emnlp-main.490/
DOI:
10.18653/v1/2024.emnlp-main.490
Bibkey:
Cite (ACL):
Revanth Gangi Reddy, Omar Attia, Yunyao Li, Heng Ji, and Saloni Potdar. 2024. AGRaME: Any-Granularity Ranking with Multi-Vector Embeddings. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 8630–8641, Miami, Florida, USA. Association for Computational Linguistics.
Cite (Informal):
AGRaME: Any-Granularity Ranking with Multi-Vector Embeddings (Gangi Reddy et al., EMNLP 2024)
Copy Citation:
PDF:
https://preview.aclanthology.org/build-pipeline-with-new-library/2024.emnlp-main.490.pdf