cs.AISep 25, 2026

LogicTree-RAG: Logic Tree-guided Retrieval-Augmented Generation for Long-form Patent Drafting

Authors: Jiaqi Zhu, Naili Xing, Hexiang Pan, Haotian Gao, Jianwei Yin, Xiaokui Xiao, Beng Chin Ooi

Organizations: National University of Singapore · Zhejiang University

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Long-form technical text generation underpins knowledge-intensive workflows, yet remains challenging for large language models (LLMs) due to the need for globally consistent logical structuring and faithful technical reasoning beyond local coherence. Patent drafting is a canonical instance of this challenge, demanding holistic generation of a legally compliant and technically exhaustive document through sustained multi-expert collaboration. Existing approaches often focus on partial section generation or rely on manually crafted outlines, limiting scalable automation in realistic settings. In this work, we propose LogicTree-RAG, a logic tree-guided retrieval-augmented generation framework that induces a hierarchical logic tree as a global organizational backbone to organize and ground technical disclosures, without relying on expert-defined drafting priors. Each node in the logic tree represents a technical element and is constructed through evidence-guided recursive generation. A hybrid traversal mechanism then maps the logic tree into patent sections, enabling controllable and section-balanced generation. Extensive experiments show that LogicTree-RAG consistently improves content quality and language conformity over strong LLM-based baselines and achieves longer structured generation with high token efficiency, demonstrating the effectiveness of logic-centric generation for complex technical document drafting.

Figures & tables

Figure 1: Architecture of LogicTree-RAG framework for long-form patent drafting.
Figure 2: Generation without and with node-aware semantic search.
Table 1: Overall performance comparison of long-form patent drafting. The best results are highlighted in bold , and the second-best results are in bold .
Figure 4
Figure 5: Section-level performance in terms of output length, content coverage, and linguistic style.
Figure 6
Figure 7: Hyperparameter Sensitivity Analysis.
Appendix figures & tables9 assets

Supplementary material from the paper’s appendix.

Appendix

NotationDescription
(X,P)(X,P)A patent-paper pair, where XX is the source (pre-publication) research paper and PP is the corresponding patent document.
X={x1,x2,…,xm}X=\{x_{1},x_{2},\dots,x_{m}\}Source paper represented as a collection of mm text spans.
P={p1,…,pn}P=\{p_{1},\dots,p_{n}\}Patent document represented as a collection of nn text spans.
X~={x~1,x~2,…,x~L}\tilde{X}=\{\tilde{x}_{1},\tilde{x}_{2},\dots,\tilde{x}_{L}\}Semantic chunks produced from XX after layout-aware semantic chunking, with L≤mL\leq m .
T=(V,E,r)\mathcal{T}=(V,E,r)Logic tree, a directed rooted tree with node set VV , edge set E⊆V×VE\subseteq V\times V , and root node rr .
V={v1,⋯ ,vN}V=\{v_{1},\cdots,v_{N}\}Set of logic tree nodes, each representing a technical element (e.g., principle, step, subsystem).
Appendix
Table 3: Notations.
MethodOutput TokensContent-levelLanguage-levelTime (s) / Patent (Token)
Coverage ↑\uparrowFactuality ↑\uparrowSemantic Similarity ↑  \uparrow\,\,Style ↑\uparrowRepetition Rate Rr\mathcal{R}_{r} ↓\downarrowCoherence ↑\uparrow
FPat\mathcal{F}_{Pat}FSrc\mathcal{F}_{Src}
Heuristic Skyline
Reference Patent (GT)18.16k89.0588.7488.8010010013.91100-
Source Paper8.03k44.7746.0688.6839.3739.268.3998.60-
Outline (Long)1.43k39.1161.7961.9310.1623.2220.1485.69-
Appendix
Table 4: Performance of LogicTree-RAG with different backbone LLMs.
Table 5: Performance comparison of LogicTree-RAG against agentic baselines for patent drafting.
MethodFully (%)Partial (%)Unsupported (%)
LLM-as-judge
COPGEN70.420.29.4
LogicTree-RAG85.612.02.4
Human Audit
COPGEN603010
LogicTree-RAG75250
Appendix
Table 6: Evidence-support analysis of retrieved evidence and generated content.
Computational Cost (/patent)Cost Breakdown
LLM Calls: 20chunking&tree init 5%, tree construction 45%, traversal 50%
Runtime: 311schunking&tree init 16%, tree construction 39%, traversal 45%
Total Input Tokens: 7.60kchunking&tree init 29%, tree construction 37%, traversal 34%
Appendix
Table 7: Supplementary computational cost analysis.
MethodOverallLegalityClaim StructureAntecedent BasisFidelity
Ours vs Qwen3-80B75.769.884.684.888.5
Ours vs DeepseekV384.465.188.590.190.2
Ours vs LongWriter78.876.590.692.085.9
Ours vs COPGEN70.565.785.282.178.5
Appendix
Table 8: Win-rate results of LogicTree-RAG in human expert evaluation.
Figure 8: Sensitivity to tree growth budget in LogicTree-RAG.
Figure 9: Qualitative case study illustrating a patent example generated by LogicTree-RAG.
Figure 10: Qualitative case study illustrating the ground-truth reference patent.
More in this topic

Long-Form Document Generation

Long-form document generation studies methods for producing coherent, factual, well-structured books, reports, and other extended texts with AI.

See all 32 papers

Explore similar work

CardsList
Aug 21, 2026

Benchmarking Patent Drafting from Inventor-Style Disclosures

While recent large language models (LLMs) have achieved promising results on individual patent drafting tasks, they fundamentally fail to investigate the core challenge of real-world patent drafting: generating a complete and legally coherent patent application directly from early-stage invention materials. Prior work predominantly assumes later-stage, highly structured, or already legalistic inputs. However, real patenting workflows begin with informal, de-legalized disclosures authored by inventors. To bridge the gap, we introduce Dis2Pat, a disclosure-to-patent dataset that reflects realistic patenting workflows by requiring the generation of complete patent applications directly from inventor-style, de-legalized disclosures. Given the inherent difficulty of long-form, legally constrained patent drafting and the strong privacy requirements, we further propose a strong baseline named Patent-MAF. It is a multi-agent framework for locally deployable patent drafting. Benchmark results reveal that current LLMs exhibit limitations in patent drafting, while Patent-MAF provides a strong baseline that consistently outperforms evaluated open-source models and remains competitive with large closed-source models.

Lekang Jiang, Wenjun Sun, Stephan Goetz

University of Cambridge · National Science Library, Chinese Academy of Sciences · Department of Information Resources Management, School of Economics and Management, University of Chinese Academy of Sciences

Language Model Generation Evaluation
AskPDF
Aug 11, 2026

Self-Knowledge Retrieval Augmented Generation Framework for Patent Matching

Patent retrieval and matching based on large language models (LLMs) play a vital role in intellectual property protection. However, due to the complex structure of patent documents, dense technical terminology, and multi-modal information, traditional methods struggle to accurately identify subtle differences between patents. Existing LLM-based patent matching approaches typically rely on domain-specific pretrained or instruction tuning, which often entail high manual labeling costs and catastrophic forgetting. While retrieval-augmented generation (RAG) methods introduce external knowledge they fail to fully leverage LLM's capability to automatically parse patents and mine deep semantic relationships. To address these limitations, this paper proposes a self-knowledge RAG framework that guides LLMs to autonomously extract key technical entities and construct hierarchical ontological structures from patent matching queries, thereby enabling query expansion and precise retrieval. The method integrates the FAISS retrieval with a generative matching mechanism, leveraging self-knowledge to enhance the model's understanding of patent innovations and significantly improve retrieval and matching accuracy. Experimental results demonstrate the outstanding performance of the proposed method on real-world patent datasets, validating its effectiveness and application potential.

Jian Zhang, Songlin Lei, Zhuohao Yang +6

School of Computer Science and Technology, Zhejiang University, Hangzhou, China · ZJU-UIUC Institute, Zhejiang University, Haining, China · Shaoxing K3i Technology Co. Ltd +1

Retrieval-Augmented GenerationInformation Retrieval
AskPDF
Jul 11, 2026

When Reasoning Hurts Legal Drafting: The Verbalization Bottleneck in Patent Claim Generation

Patent claim drafting is a challenging legal drafting task that requires technical expertise, precise linguistic control, strict adherence to formal conventions, and the preservation of complex logical relationships among claim elements. While Chain-of-Thought (CoT) prompting has been widely used to improve the reasoning capabilities of large language models (LLMs), recent evidence suggests that its benefits may be limited, or even negative, in highly structured or pattern-sensitive tasks. Therefore, this paper investigates whether CoT prompting benefits patent claim generation. We propose a task-specific CoT method for patent claim generation and evaluate its effectiveness through both automatic metrics and human expert assessment. Our results show that reasoning-enhanced prompting can improve claim quality. Moreover, we demonstrate a counter-intuitive but important empirical finding: implicit CoT, where reasoning is kept internal rather than explicitly verbalized, consistently outperforms explicit CoT. Through systematic analysis, we show that explicit CoT can introduce an unnecessary information bottleneck for claim generation. Verbalized reasoning may compromise the quality of final outputs through three specific mechanisms: abstraction of critical details, disruption of internalized generation patterns, and cascading error propagation. Our findings provide new insights into legal tasks and CoT applications.

Lekang Jiang, Wenjun Sun, Stephan Goetz

University of Cambridge · National Science Library, Chinese Academy of Sciences; Department of Information Resources Management, School of Economics and Management, University of Chinese Academy of Sciences.

LLM PromptingCoT ReasoningText GenerationLLM Reasoning
AskPDF