Metraipolis › archive › dc_2447bfa46486
Chunk Recomputation under Real Document Revisions: Measuring Update Cost and Retrieval Quality of Fixed, Recursive, and Content-Defined Chunking — arXiv:2610.31019
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.31019
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-23T15:20:38.203Z
- SHA-256 of the archived text
ebafae9ec1742e7dd8f2719e980e8f291ef807250e90e8c2fa162e0d6c81a35e- Stable URL of this page
https://metraipolis.com/documento/dc_2447bfa46486
How to cite
Chunk Recomputation under Real Document Revisions: Measuring Update Cost and Retrieval Quality of Fixed, Recursive, and Content-Defined Chunking — arXiv:2610.31019 — arXiv (en), https://arxiv.org/abs/2610.31019 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2447bfa46486 (Metraipolis archive dc_2447bfa46486, sha256 ebafae9ec1742e7dd8f2719e980e8f291ef807250e90e8c2fa162e0d6c81a35e)
BibTeX
@misc{dc2447bfa46486,
title = {Chunk Recomputation under Real Document Revisions: Measuring Update Cost and Retrieval Quality of Fixed, Recursive, and Content-Defined Chunking — arXiv:2610.31019},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.31019},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2447bfa46486; sha256 ebafae9ec1742e7dd8f2719e980e8f291ef807250e90e8c2fa162e0d6c81a35e},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.31019. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Chunking strategies for retrieval-augmented generation are almost always evaluated on a static snapshot, yet the documents underlying a knowledge base continue to change. Because chunks are embedded independently, a boundary that moves after an edit forces unchanged text to be re-embedded. We measure this update cost on 120 pairs of adjacent Chinese and English Wikipedia revisions for three stateless chunkers (non-overlapping fixed-size, recursive paragraph-first, and content-defined chunking) under a strict byte-identical reuse rule and a frozen dense encoder. On the 116-article common-success pool, with frozen, length-calibrated parameters, fixed-size chunking recomputes 57-68% of its full new-version embedding-input workload, recursive chunking 38-56%, and content-defined chunking 43-54% (it failed on 4 of 62 Chinese articles), while at these parameters mean evidence recall under a fixed 2048-token retrieval budget differs by less than two percentage points within either language. A cost-only parameter grid shows that the ranking of the two content-aware chunkers changes with chunk-length settings, whereas non-overlapping fixed-size chunking has the highest recomputation cost in every length-matched condition. Controlled edits and a verbatim-presence diagnostic support boundary drift as one contributor to recomputation of text already present in the previous version. Update cost is therefore a substantial, parameter-sensitive dimension that static retrieval evaluations do not expose and that should be reported alongside retrieval quality. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno). Categorie arXiv: Computation and Language (cs.CL) ; Information Retrieval (cs.IR). Pagina del lavoro: https://arxiv.org/abs/2610.31019 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-23T15:09:38.347Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).