Chunk Recomputation under Real Document Revisions: Measuring Update Cost and Retrieval Quality of Fixed, Recursive, and Content-Defined Chunking

A study of 120 pairs of adjacent Chinese and English Wikipedia revisions found that fixed-size chunking recomputes 57-68% of its full new-version embedding-input workload after edits, versus 38-56% for recursive paragraph-first chunking and 43-54% for content-defined chunking, under a strict byte-identical reuse rule with a frozen dense encoder. On the 116-article common-success pool, mean evidence recall under a fixed 2048-token retrieval budget differed by less than two percentage points within either language, and content-defined chunking failed on 4 of 62 Chinese articles. The authors, posting as arXiv:2610.31019v1, conclude that update cost is a parameter-sensitive dimension static retrieval evaluations do not expose and should be reported alongside retrieval quality.

arXiv:2610.31019v1 Announce Type: new Abstract: 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.