Succeeding at Scale: Automated Dataset Construction and Query-Side Adaptation for Multi-Tenant Search

TL;DR

This paper presents DevRev-Search, a fully automated benchmark for technical support retrieval, and introduces an index-preserving adaptation method that fine-tunes only the query encoder to improve multi-tenant search systems efficiently.

Contribution

It introduces a novel automated dataset construction pipeline and a parameter-efficient fine-tuning strategy for scalable multi-tenant search adaptation.

Findings

PEFT of query encoder improves retrieval quality significantly.

Index-preserving adaptation reduces computational costs.

Automated benchmark enables scalable evaluation.

Abstract

Large-scale multi-tenant retrieval systems generate extensive query logs but lack curated relevance labels for effective domain adaptation, resulting in substantial underutilized "dark data". This challenge is compounded by the high cost of model updates, as jointly fine-tuning query and document encoders requires full corpus re-indexing, which is impractical in multi-tenant settings with thousands of isolated indices. We introduce DevRev-Search, a passage retrieval benchmark for technical customer support built via a fully automated pipeline. Candidate generation uses fusion across diverse sparse and dense retrievers, followed by an LLM-as-a-Judge for consistency filtering and relevance labeling. We further propose an Index-Preserving Adaptation strategy that fine-tunes only the query encoder, achieving strong performance gains while keeping document indices fixed. Experiments on DevRev-Search, SciFact, and FiQA-2018 show that Parameter-Efficient Fine-Tuning (PEFT) of the query encoder delivers a remarkable quality-efficiency trade-off, enabling scalable and practical enterprise search adaptation.