QESChunker: A Single Objective Unifies Overlapping and Non-Overlapping Chunking for RAG
Abstract
Chunking is a fundamental preprocessing step in retrieval-augmented generation (RAG), yet most existing chunkers rely primarily on heuristics. We introduce the Question--Evidence-Supervised Chunker (QESChunker), to our knowledge the first method for constructing static chunk indices from question--evidence supervision. QESChunker constructs document-specific chunk indices by optimizing over human-annotated or LLM-generated question--evidence pairs, without access to held-out test queries. We formulate non-overlapping and overlapping index construction through a unified question--evidence utility-maximization objective, turning chunking into a supervised combinatorial optimization problem. Under the non-overlap constraint, the objective decomposes over chunks and can be optimized exactly in polynomial time by dynamic programming. When overlap is allowed, the objective becomes monotone submodular; exact optimization is NP-hard, but our greedy achieves a (1−1/√e)-approximation. Across two self-constructed and three public datasets, both variants outperform existing chunking methods in complete-evidence recall in nearly all evaluated settings, under the same overlap regime and a fixed hybrid retriever. These results recast chunking as supervised, budget-aware index construction for evidence retrieval.
Related
- Simple Extensions of Single-Objective Acquisition Functions and Hedge Strategies for Multi-Objective Bayesian Optimization
- Weak Tiling by Unions of Non-overlapping Unit Cubes
- Bayesian Reconciliation of Overlapping Multi-Step Forecasts: A Post-Processing Framework for Cross-Origin Coherence
- Resource-Constrained Semantic-Aware Remote Estimation with Overlapping Sensor Coverage
- Hybrid coupling with numerics-informed neural networks and the overlapping Schwarz alternating method
BibTeX
@misc{zhao2026qeschunker,
title = {{QESChunker: A Single Objective Unifies Overlapping and Non-Overlapping Chunking for RAG}},
author = {Yifan Zhao and Qianyang Zhao and Xinyu Guan and kai wei and Yuming Deng},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.36557},
eprint = {2610.36557},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.36557},
}
RIS
TY - UNPB
TI - QESChunker: A Single Objective Unifies Overlapping and Non-Overlapping Chunking for RAG
AU - Yifan Zhao
AU - Qianyang Zhao
AU - Xinyu Guan
AU - kai wei
AU - Yuming Deng
PY - 2026
DA - 2026/10/27
PB - arXiv
DO - 10.48550/arxiv.2610.36557
UR - https://arxiv.org/abs/2610.36557
AB - Chunking is a fundamental preprocessing step in retrieval-augmented generation (RAG), yet most existing chunkers rely primarily on heuristics. We introduce the Question--Evidence-Supervised Chunker (QESChunker), to our knowledge the first method for constructing static chunk indices from question--evidence supervision. QESChunker constructs document-specific chunk indices by optimizing over human-annotated or LLM-generated question--evidence pairs, without access to held-out test queries. We formulate non-overlapping and overlapping index construction through a unified question--evidence utility-maximization objective, turning chunking into a supervised combinatorial optimization problem. Under the non-overlap constraint, the objective decomposes over chunks and can be optimized exactly in polynomial time by dynamic programming. When overlap is allowed, the objective becomes monotone submodular; exact optimization is NP-hard, but our greedy achieves a (1−1/√e)-approximation. Across two self-constructed and three public datasets, both variants outperform existing chunking methods in complete-evidence recall in nearly all evaluated settings, under the same overlap regime and a fixed hybrid retriever. These results recast chunking as supervised, budget-aware index construction for evidence retrieval.
ER -