cs.IRNov 17, 2026

Teaching Language Models to Narrow Boolean Patent Searches from Search Feedback

Authors: Seiya Kawano, Koji Marusaki, Hirofumi Nonaka, Koichiro Yoshino

Organizations: Kyoto Institute of Technology, Kyoto, Japan · Guardian Robot Project, RIKEN, Kyoto, Japan · Nagaoka University of Technology, Nagaoka, Japan · Aichi Institute of Technology, Nagoya, Japan · 

Abstract

Patent prior-art searchers issue Boolean queries line by line, broadening to find relevant documents and narrowing until candidates can be screened. We distill 38,397 Japan Patent Office search reports into executable expert strategies, re-execute them, and evaluate generated strategies by citation recall and the smallest set containing each citation. Prompted models write valid, high-recall strategies but leave citations in sets too large to screen. We adapt a generator offline from scored rollouts in two forms: strategy-level learning from a screening-aware strategy score, and line-level learning from citation-preserving narrowing credit, which upweights the query lines keeping a found citation in a smaller set. On shared rollouts, line-level credit outperforms strategy-level scores on screening measures; iterated, it cuts the median citation-containing set from 768 to under 50 and places more citations in sets under 100 documents than experts, at lower recall. Strategy-level preference learning reaches near-complete recall; the two forms are complementary.

Figures & tables

Explore similar work

CardsList
  1. Citation-Driven Multi-View Training for Patent Embeddings: QaECTER and Sophia-Bench

    Apr 24, 2026Younes Djemmal, You Zuo, Kim Gerdes +1Legal IRText Embedding Models

  2. From Rules to Neural Graphs: Scalable Structured Prediction for Patent Prior Art Search

    Oct 1, 2026Nikolai Zenovkin, Sebastian BjörkqvistLegal IRGraph-Based Retrieval

  3. PatentAgentBench: Benchmarking LLM Agents with Professional Patent Task Suites

    Oct 26, 2026Yijie Hu, Wei Li, Junjie Zhou +6LLM Agent EvaluationAI Agent Benchmarks