Metraipolis › archive › dc_c1406dc93e99
Tree Navigation Without LLM Summaries: A Matched-Cost Study of Hierarchical Retrieval for Long-Document QA — arXiv:2610.06902
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.06902
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T11:22:56.228Z
- SHA-256 of the archived text
0d6e9691e279f4edcaf3d98ff61e63253484108f4556c4ea82d2ce30a4741462- Stable URL of this page
https://metraipolis.com/documento/dc_c1406dc93e99
How to cite
Tree Navigation Without LLM Summaries: A Matched-Cost Study of Hierarchical Retrieval for Long-Document QA — arXiv:2610.06902 — arXiv (en), https://arxiv.org/abs/2610.06902 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c1406dc93e99 (Metraipolis archive dc_c1406dc93e99, sha256 0d6e9691e279f4edcaf3d98ff61e63253484108f4556c4ea82d2ce30a4741462)
BibTeX
@misc{dcc1406dc93e99,
title = {Tree Navigation Without LLM Summaries: A Matched-Cost Study of Hierarchical Retrieval for Long-Document QA — arXiv:2610.06902},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.06902},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c1406dc93e99; sha256 0d6e9691e279f4edcaf3d98ff61e63253484108f4556c4ea82d2ce30a4741462},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.06902. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Retrieval-augmented generation grounds language models in external context, but for long documents flat top-$k$ retrieval can cluster on a single region and miss complementary evidence. RAPTOR-style summary trees address this by recursively clustering chunks and using a language model to summarize each cluster at indexing time, then ranking summary nodes alongside raw chunks at query time. We show the main benefit of summary trees in long-document QA can come from navigation rather than the generated summary content. We introduce NavTree, a leaves-only retriever that builds a deterministic balanced segment tree over chunks (zero language-model calls at indexing) and uses the tree purely as a navigation scaffold: a hybrid lexical-and-dense frontier walk, anchored on top retrieved leaves, descends from the root and emits only leaf chunks to the reader. On a matched-cost evaluation against flat retrievers and an extractive re-implementation of RAPTOR, NavTree is the strongest matched-cost hierarchical retriever in our evaluated grid and ties the strongest flat baseline. On long-document multi-hop QA, it is the only hierarchical method that significantly beats BM25 on a class-vs-class basis, corroborated by a reader-free retrieval-recall check. A matched-reader replication of the published abstractive RAPTOR variant, given strong cluster summaries, still loses to NavTree at every multi-chunk budget, at zero indexing cost. The ranking carries across stronger and open-weight readers, a stronger encoder, and a full factorial that isolates leaves-only emission as the structural lever. 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/2610.06902 — 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 2026-10-07T11:15:39.404Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).