ORDER: Task-Conditioned Routing for Retrieval-Augmented Generation

arXiv cs.AI Papers
retrieval-augmented-generation query-routing indexing-adaptation retrieval-systems machine-learning academic-research

Summary

ORDER introduces a framework that dynamically adapts indexing and retrieval strategies in RAG systems based on the query, improving performance in expert domains.

arXiv:2609.17012v1 Announce Type: new Abstract: Retrieval-Augmented Generation (RAG) pipelines typically rely on a fixed indexing and retrieval configuration determined at preprocessing time. This one-size-fits-all design is ill-suited to domain-expert settings, where heterogeneous queries require different chunking granularities, metadata constraints, and source-selection strategies. As a result, configurations that are effective for one family of queries often perform poorly for others. In this paper, we introduce ORDER (Optimal Routing for Dynamic Evidence Retrieval), a query-conditioned RAG framework that jointly adapts indexing and retrieval to the incoming query. Our approach first discovers semantic clusters over a given set of questions associated to a corpus and learns, for each cluster, a chunking strategy together with a suited metadata filtering and reranking configuration. At inference time, queries are routed to the appropriate pre-built index through nearest-centroid assignment. To further improve retrieval, we propose a supervised query router (QRe) that predicts which collections are most likely to contain relevant evidence, coupled with a Uniform Multi-source Sampler (UMS) that allocates the retrieval budget evenly across the selected sources. We evaluate our framework on large-scale, heterogeneous historical archives and show that conditioning both indexing and retrieval on the query consistently outperforms both naive baselines and strong state-of-the-art RAG systems in complex expert-domain environments.
Original Article
View Cached Full Text

Cached at: 09/16/26, 09:06 AM 

# ORDER: Task-Conditioned Routing for Retrieval-Augmented Generation
Source: [https://arxiv.org/abs/2609.17012](https://arxiv.org/abs/2609.17012)
[View PDF](https://arxiv.org/pdf/2609.17012)

> Abstract:Retrieval\-Augmented Generation \(RAG\) pipelines typically rely on a fixed indexing and retrieval configuration determined at preprocessing time\. This one\-size\-fits\-all design is ill\-suited to domain\-expert settings, where heterogeneous queries require different chunking granularities, metadata constraints, and source\-selection strategies\. As a result, configurations that are effective for one family of queries often perform poorly for others\. In this paper, we introduce ORDER \(Optimal Routing for Dynamic Evidence Retrieval\), a query\-conditioned RAG framework that jointly adapts indexing and retrieval to the incoming query\. Our approach first discovers semantic clusters over a given set of questions associated to a corpus and learns, for each cluster, a chunking strategy together with a suited metadata filtering and reranking configuration\. At inference time, queries are routed to the appropriate pre\-built index through nearest\-centroid assignment\. To further improve retrieval, we propose a supervised query router \(QRe\) that predicts which collections are most likely to contain relevant evidence, coupled with a Uniform Multi\-source Sampler \(UMS\) that allocates the retrieval budget evenly across the selected sources\. We evaluate our framework on large\-scale, heterogeneous historical archives and show that conditioning both indexing and retrieval on the query consistently outperforms both naive baselines and strong state\-of\-the\-art RAG systems in complex expert\-domain environments\.

## Submission history

From: Aurelien PELLET \[[view email](https://arxiv.org/show-email/0e25dc32/2609.17012)\] \[via CCSD proxy\] **\[v1\]**Tue, 15 Sep 2026 11:23:16 UTC \(180 KB\)

Similar Articles

LightRAG: Simple and Fast Retrieval-Augmented Generation

Papers with Code Trending

The article introduces LightRAG, an open-source framework that enhances Retrieval-Augmented Generation by integrating graph structures for improved contextual awareness and efficient information retrieval.

Query-Aware Source-Risk Triage for Retrieval-Augmented Generation

arXiv cs.AI

The paper proposes a query-aware triage layer for retrieval-augmented generation (RAG) pipelines to handle the material relationship of sources to queries, using scoring and routing methods evaluated on synthetic data to define risk-coverage targets.

Candidate-Constrained Retrieval-Augmented Generation for LongEval-RAG: System Design and Empirical Analysis

arXiv cs.CL

This paper presents a candidate-constrained RAG system for the LongEval-RAG task at CLEF 2026, combining deterministic provenance tracking with passage retrieval, query expansion, pseudo-relevance feedback, reciprocal rank fusion, evidence reranking, and citation-aware aggregation. An ablation study of ten pipeline variants shows that a rule-based chunking pipeline with sentence-level neural selection achieves the best performance.