Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

From RAG to Riches: Retrieval Interlaced with Sequence Generation

Palak Jain, Livio Baldini Soares, Tom Kwiatkowski


Abstract
We present RICHES, a novel approach that interleaves retrieval with sequence generation tasks. RICHES offers an alternative to conventional RAG systems by eliminating the need for separate retriever and generator. It retrieves documents by directly decoding their contents, constrained on the corpus. Unifying retrieval with generation allows us to adapt to diverse new tasks via prompting alone. RICHES can work with any Instruction-tuned model, without additional training. It provides attributed evidence, supports multi-hop retrievals and interleaves thoughts to plan on what to retrieve next, all within a single decoding pass of the LLM. We demonstrate the strong performance of RICHES across ODQA tasks including attributed and multi-hop QA.
Anthology ID:
2024.emnlp-main.502
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:
8887–8904
Language:
URL:
https://aclanthology.org/2024.emnlp-main.502/
DOI:
10.18653/v1/2024.emnlp-main.502
Bibkey:
Cite (ACL):
Palak Jain, Livio Baldini Soares, and Tom Kwiatkowski. 2024. From RAG to Riches: Retrieval Interlaced with Sequence Generation. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 8887–8904, Miami, Florida, USA. Association for Computational Linguistics.
Cite (Informal):
From RAG to Riches: Retrieval Interlaced with Sequence Generation (Jain et al., EMNLP 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.emnlp-main.502.pdf