Splitting an agent skill into its name, description, and body during retrieval, then learning how to weight per-field similarity scores, outperforms flat concatenation on SkillRet and SRA-Bench and gains more as the skill bank grows.
Integrating the Probabilistic Models BM25/BM25F into Lucene
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
This document describes the BM25 and BM25F implementation using the Lucene Java Framework. Both models have stood out at TREC by their performance and are considered as state-of-the-art in the IR community. BM25 is applied to retrieval on plain text documents, that is for documents that do not contain fields, while BM25F is applied to documents with structure.
fields
cs.IR 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Field Aware Agent Skill Retrieval
Splitting an agent skill into its name, description, and body during retrieval, then learning how to weight per-field similarity scores, outperforms flat concatenation on SkillRet and SRA-Bench and gains more as the skill bank grows.