Fragment-Level Evidence Retrieval for Complex Long Texts in the Power Domain via Recursive Character Splitting
Year: 2026
Type: article
Abstract: Knowledge bases for primary power equipment must retrieve concise, verifiable evidence from long standards, fault cases, and maintenance records. We evaluate six retrieval-unit strategies, three retrievers, and two Chinese embedding encoders using 149 documents (1.73 million characters) and 48 author-constructed questions, with 20 for development and 28 held out for testing. At the matched 500-character/100-overlap setting, recursive splitting preserves 84.5% of numbered standards clauses, compared with 78.6% for sliding-window, 72.0% for paragraph, and 64.9% for semantic chunking; clause-aware splitting reaches 88.3%. Recursive splitting with dense retrieval records Hit@10 of 0.571 and evidence coverage@10 of 0.542; hybrid dense–BM25 fusion raises these to 0.714 and 0.685. Sliding-window dense retrieval records 0.786 and 0.685 on the same measures. Recursive, sliding-window, paragraph, and semantic chunks each contain all 112 annotated evidence spans, whose lengths range from 11 to 90 characters (median 36); clause-aware chunking contains 97.3%. The results characterize the structural-preservation and retrieval profiles of the tested strategies and show that hybrid fusion improves top-10 evidence recovery with recursive chunks.
Source: Electronics
Authors: Xiawei Cheng(Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing), Wanden Mao(Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing), Joshua Luhwago(Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing), Asad Ullah(Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing), Guojing Qian(Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing), Shengzhe Xi(Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing), Shanfeng Liu(Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing), Li Ruan(Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing), Limin Xiao(Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing)
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Keywords: evidence retrieval, dense retrieval, BM25
Open Access status: gold
Estimated APC paid: $2,896
OpenAlex ID: W7236275542
