Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2604.20849 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908987669086208 |
|---|---|
| author | Rainey, Mike Acar, Umut Sezer, Muhammed |
| author_facet | Rainey, Mike Acar, Umut Sezer, Muhammed |
| contents | Retrieval-augmented generation over semi-structured sources such as HTML is constrained by a mismatch between document structure and the flat, sequence-based interfaces of today's embedding and generative models. Retrieval pipelines often linearize documents into fixed-size chunks before indexing, which obscures section structure, lists, and tables, and makes it difficult to return small, citation-ready evidence without losing the surrounding context that makes it interpretable.
We present a structure-aware retrieval pipeline that operates over tree-structured documents. The core idea is to represent candidates as subdocuments: precise, addressable selections that preserve structural identity while deferring the choice of surrounding context. We define a small set of document primitives--paths and path sets, subdocument extraction by pruning, and two contextualization mechanisms. Global contextualization adds the non-local scaffolding needed to make a selection intelligible (e.g., titles, headers, list and table structure). Local contextualization expands a seed selection within its structural neighborhood to obtain a compact, context-rich view under a target budget. Building on these primitives, we describe an embedding-based candidate generator that indexes sentence-seeded subdocuments and a query-time, document-aware aggregation step that amortizes shared structural context. We then introduce a contextual filtering stage that re-scores retrieved candidates using locally contextualized views.
Across experiments on HTML question-answering benchmarks, we find that preserving structure while contextualizing selections yields higher-quality, more diverse citations under fixed budgets than strong passage-based baselines, while maintaining scalability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_20849 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | SPIRE: Structure-Preserving Interpretable Retrieval of Evidence Rainey, Mike Acar, Umut Sezer, Muhammed Information Retrieval Artificial Intelligence Computation and Language Retrieval-augmented generation over semi-structured sources such as HTML is constrained by a mismatch between document structure and the flat, sequence-based interfaces of today's embedding and generative models. Retrieval pipelines often linearize documents into fixed-size chunks before indexing, which obscures section structure, lists, and tables, and makes it difficult to return small, citation-ready evidence without losing the surrounding context that makes it interpretable. We present a structure-aware retrieval pipeline that operates over tree-structured documents. The core idea is to represent candidates as subdocuments: precise, addressable selections that preserve structural identity while deferring the choice of surrounding context. We define a small set of document primitives--paths and path sets, subdocument extraction by pruning, and two contextualization mechanisms. Global contextualization adds the non-local scaffolding needed to make a selection intelligible (e.g., titles, headers, list and table structure). Local contextualization expands a seed selection within its structural neighborhood to obtain a compact, context-rich view under a target budget. Building on these primitives, we describe an embedding-based candidate generator that indexes sentence-seeded subdocuments and a query-time, document-aware aggregation step that amortizes shared structural context. We then introduce a contextual filtering stage that re-scores retrieved candidates using locally contextualized views. Across experiments on HTML question-answering benchmarks, we find that preserving structure while contextualizing selections yields higher-quality, more diverse citations under fixed budgets than strong passage-based baselines, while maintaining scalability. |
| title | SPIRE: Structure-Preserving Interpretable Retrieval of Evidence |
| topic | Information Retrieval Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2604.20849 |