Contextualization with SPLADE for High Recall Retrieval

TL;DR

This paper introduces a novel approach combining SPLADE with linear models for high recall retrieval, significantly reducing review costs in eDiscovery and medical review tasks by leveraging contextualized sparse vectors.

Contribution

It proposes a method that integrates SPLADE's contextualization with linear models to improve efficiency and effectiveness in high recall retrieval tasks.

Findings

Reduces review costs by 10% and 18% in two evaluation collections.

Leverages both pretrained language models and linear models for better retrieval.

Applicable in one-phase review workflows with 80% recall target.

Abstract

High Recall Retrieval (HRR), such as eDiscovery and medical systematic review, is a search problem that optimizes the cost of retrieving most relevant documents in a given collection. Iterative approaches, such as iterative relevance feedback and uncertainty sampling, are shown to be effective under various operational scenarios. Despite neural models demonstrating success in other text-related tasks, linear models such as logistic regression, in general, are still more effective and efficient in HRR since the model is trained and retrieves documents from the same fixed collection. In this work, we leverage SPLADE, an efficient retrieval model that transforms documents into contextualized sparse vectors, for HRR. Our approach combines the best of both worlds, leveraging both the contextualization from pretrained language models and the efficiency of linear models. It reduces 10% and 18% of the review cost in two HRR evaluation collections under a one-phase review workflow with a target recall of 80%. The experiment is implemented with TARexp and is available at https://github.com/eugene-yang/LSR-for-TAR.