PatRe: A Full-Stage Office Action and Rebuttal Generation Benchmark for Patent Examination

2026-05-05

AI Summary

Existing patent examination benchmarks are largely confined to static classification or information extraction, failing to capture the dynamic, iterative dialogue between examiners and applicants. This work proposes PatRe, the first interactive benchmark encompassing the full patent examination lifecycle, comprising 480 real-world cases. PatRe uniquely frames patent examination as a multi-turn argumentation and response task, enabling the generation of both examiner objections and applicant rebuttals. It introduces a dual evaluation paradigm grounded in authentic context and retrieval-augmented reasoning. Experiments reveal a substantial performance gap between closed- and open-source large language models in legal reasoning and novelty assessment, and uncover an asymmetry in task difficulty between the examiner and applicant perspectives, offering a new benchmark and critical insights for intelligent patent examination research.

Abstract

Patent examination is a complex, multi-stage process requiring both technical expertise and legal reasoning, increasingly challenged by rising application volumes. Prior benchmarks predominantly view patent examination as discriminative classification or static extraction, failing to capture its inherently interactive and iterative nature, similar to the peer review and rebuttal process in academic publishing. In this paper, we introduce PatRe, the first benchmark that models the full patent examination lifecycle, including Office Action generation and applicant rebuttal. PatRe comprises 480 real-world cases and supports both oracle and retrieval-simulated evaluation settings. Our benchmark reframes patent examination as a dynamic, multi-turn process of justification and response. Extensive experiments across various LLMs reveal critical insights into model performance, including differences between proprietary and open-source models, as well as task asymmetries between examiner analysis and applicant-side rebuttal. These findings highlight both the potential and current limitations of LLMs in modeling complex, real-world legal reasoning and technical novelty judgment in patent examination. We release our code and dataset to facilitate future research on patent examination modeling.

Problem

• patent examination
• Office Action
• rebuttal generation
• interactive process
• legal reasoning

Innovation

• patent examination
• Office Action generation
• applicant rebuttal
• multi-turn legal reasoning
• LLM benchmark