mlsanityData to 7 Oct 2026

Structure-Aware Semantic Chunking with Title-Chain Prefixes: A 1600-Query Evaluation and the Measurement Trap in Text-Transform Ablations

arXiv 2608.00824 ↗First posted 1 Aug 2026cs.SEcited by 0

On the map: Retrieval-augmented generation → Multimodal RAG

Abstract (arXiv)

Chunking is the first and most consequential step in retrieval-augmented generation (RAG): every downstream retrieval decision inherits the chunk boundaries. We present a three-stage, chunk-side-only semantic chunking pipeline---header-split, semantic merge, and title-chain prefixing---that costs zero additional LLM calls: the title chain reuses the document's own header hierarchy instead of a generated summary. On a 1600-query stratified evaluation over a production Markdown knowledge base, the pipeline improves MRR@5 from 0.374 to 0.463 (+23.8%) on the full set and from 0.828 to 0.925 (+11.7%) on the answerable subset (n=563), with dual-annotator Cohen's kappa 0.45 (unweighted, 16,000 score pairs). We then report what we tried and what failed: three query-side or architecture-level follow-ups are design dead ends (unevaluated---no comparable run artifacts), one measured failure (prefix weight decay), and one protocol-level failure that is the paper's central methodological finding. In a same-pool prefix on/off ablation, dual-annotator agreement collapsed from kappa 0.45 to 0.04 under identical prompts---stripping the title-chain context strips the disambiguation signal annotators need to agree on relevance. This measurement trap invalidates a common evaluation practice in chunking research and motivates retrieval-time per-candidate prefix evaluation, the direction we recommend from all our evidence.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Multimodal RAG, one of 15 directions in Retrieval-augmented generation. Builds on 8 of its 12 references that are in the map. First posted on arXiv 1 Aug 2026.

Builds on

8 of 12 references.

Closest ancestors · 2

Same direction, newest first

  1. 2026The Effect of Text Chunk Size on Retrieval-Augmented Generation Performance8 Jun
  2. 2025Rethinking Chunk Size For Long-Document Retrieval: A Multi-Dataset Analysis27 May

Other references · 6

  1. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  2. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,049
  3. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,779
  4. 2017Billion-scale similarity search with GPUscited by 2,477
  5. 2024RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrievalcited by 519
  6. 2022Precise Zero-Shot Dense Retrieval without Relevance Labelscited by 508

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Evaluating Chunking Strategies For Retrieval-Augmented Generation in Oil and Gas Enterprise Documents
  2. 2025A New HOPE: Domain-agnostic Automatic Evaluation of Text Chunking
  3. 2026Does Document Structure Help Dense Retrieval? A Placebo-Controlled Ablation of Four Mechanisms Across Two Corpora
  4. 2026Query-Adaptive Semantic Chunking for Retrieval-Augmented Generation: A Dynamic Strategy with Contextual Window Expansion
  5. 2026Web Retrieval-Aware Chunking (W-RAC) for Efficient and Cost-Effective Retrieval-Augmented Generation Systems
  6. 2024Is Semantic Chunking Worth the Computational Cost?
  7. 2025Breaking It Down: Domain-Aware Semantic Segmentation for Retrieval Augmented Generation
  8. 2026A Systematic Analysis of Chunking Strategies for Reliable Question Answering