De-DSI

Decentralised Differentiable Search Index

Conference Paper (2024)
Author(s)

P.M. Neague (TU Delft - Data-Intensive Systems)

Marcel Gregoriadis (TU Delft - Data-Intensive Systems)

J.A. Pouwelse (TU Delft - Data-Intensive Systems)

Research Group
Data-Intensive Systems
DOI related publication
https://doi.org/10.1145/3642970.3655837
More Info
expand_more
Publication Year
2024
Language
English
Research Group
Data-Intensive Systems
Pages (from-to)
134-143
ISBN (print)
979-8-4007-0541-0
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

This study introduces De-DSI, a novel framework that fuses large language models (LLMs) with genuine decentralization for information retrieval, particularly employing the differentiable search index (DSI) concept in a decentralized setting. Focused on efficiently connecting novel user queries with document identifiers without direct document access, De-DSI operates solely on query-docid pairs. To enhance scalability, an ensemble of DSI models is introduced, where the dataset is partitioned into smaller shards for individual model training. This approach not only maintains accuracy by reducing the number of data each model needs to handle but also facilitates scalability by aggregating outcomes from multiple models. This aggregation uses a beam search to identify top docids and applies a softmax function for score normalization, selecting documents with the highest scores for retrieval. The decentralized implementation demonstrates that retrieval success is comparable to centralized methods, with the added benefit of the possibility of distributing computational complexity across the network. This setup also allows for the retrieval of multimedia items through magnet links, eliminating the need for platforms or intermediaries.