Paper 2607.24223
A New Role for Relevance: Guiding Corpus Interaction in Agentic Search
- Published
- Jul 2026
- Research lab
- Independent
- Citations
- 0
- GitHub
- 51 stars
01 In brief
Summary
The paper introduces the Relevance-Aware RipGrep Search Agent (RARG), which uses relevance as an execution prior to guide corpus interaction in agentic search.
RARG ranks documents with an embedding retriever and makes ripgrep traverse them in that order, exposing matches from more relevant documents earlier.
It has two extensions: RARG+ seeds the agent with query-relevant paragraphs as an entry point, and RARG++ reranks grep matches to surface informative excerpts from lower-ranked documents.
On BrowseComp-Plus (100K documents), RARG++ with GPT-5.4-mini achieves 84% accuracy (vs.
78% for RISE and DCI) with 23.9 average tool calls (vs.
28.7 and 99.1).
With GPT-5.4, RARG++ reaches 91% accuracy, exceeding RISE by 9 points.
When scaling to 1M documents, RARG++ retains 79% accuracy vs.
69% for RISE.
On BRIGHT, RARG+ achieves 53.36 average nDCG@10, outperforming DCI, RISE, and NeMo.
The results show that relevance-aware interaction improves accuracy and efficiency, with faster and more reliable search convergence.
02 From the paper
Abstract
Relevance is a query-dependent estimate of whether a document or excerpt contains useful evidence. Existing retrieval agents use relevance to select top-$k$ content, but document relevance alone cannot localize, compose, or verify the evidence required by complex questions. Direct Corpus Interaction (DCI) enables such fine-grained operations through grep-style exploration, but its relevance-agnostic search can expose useful clues late and delay convergence. Recent advances use relevance to narrow the corpus into a working space for interaction. Once interaction begins, however, relevance still does not directly guide which documents grep searches first or distinguish informative excerpts from a broad set of matches to let LLMs see them first. We introduce the Relevance-Aware RipGrep Search Agent (RARG), which turns relevance into an execution prior for corpus interaction. RARG provides coarse-to-fine relevance guidance: it orders documents for sequential 'ripgrep' traversal to expose globally relevant clues earlier, initializes promising entry points with query-relevant paragraphs, and reranks grep matches to surface informative excerpts that document-level ranking may otherwise obscure. Across challenging browse question answering and reasoning-intensive retrieval, RARG improves the accuracy--efficiency frontier over retrieval-based and direct-interaction agents. These results demonstrate that relevance-aware interaction enables faster and more reliable search convergence.