Selected work

Representative Papers

Graph-Based Agentic AI with LangGraph: Workflow Pathways for Long-Running Stateful Business Processes

Jul 21, 2026

This work addresses the challenges of constructing and managing generative AI agent systems for long-horizon, stateful, multi-step business processes by proposing a graph-structured workflow design methodology. Leveraging the LangGraph framework, it explicitly models core mechanisms such as state management, conditional routing, and human-in-the-loop interventions. The approach is instantiated in three representative applications: SQL analysis with repair loops, retrieval-augmented generation gated by evidential validation, and human-AI collaborative policy review supporting interruption and checkpoint-based recovery. By treating behaviors like routing, pausing, and audit trails as explicit product features rather than implicit prompt logic, this study not only delineates the applicability boundaries of LangGraph in high-complexity workflows but also substantially enhances system controllability, reliability, and auditability in real-world operational settings, establishing a reusable engineering paradigm.

0 citationsRead paper

From Patent Expiry to Business Pathways: AI Workflows for Activating Innovation Archives

Jul 11, 2026

This study addresses the underutilization of expired or lapsed patents, whose embedded technical knowledge often remains inaccessible due to difficulties in identification and interpretation, thereby hindering innovation transfer. The work proposes the first AI-driven framework that treats patent expiration as both a commercial opportunity and a knowledge archival milestone, integrating legal status, technical content, and market signals to generate structured business pathways—such as SaaS offerings or consulting services. The methodology combines patent metadata, fee payment records, semantic search, patent family analysis, and a locally deployed Qwen3.6 generative model to enable multidimensional risk assessment. Evaluated on the CIPO dataset, the approach successfully identified 20 high-potential patents, demonstrating the stability of the scoring model and compliance of AI-generated outputs, while also uncovering practical challenges such as insufficient legal status coverage.

0 citationsRead paper

Pandas for Reproducible Data Analysis: From Spreadsheets to Research-Grade Python Workflows

Jun 12, 2026

This work addresses the significant limitations of spreadsheet-based analysis in reproducibility, auditability, version control, and automation. It proposes a migration pathway from Excel to research-grade analytical workflows by leveraging Python’s pandas library as a bridge. The study introduces an innovative set of Excel-to-pandas mapping rules, categorizes nine canonical workflow patterns, and compiles a catalog of common failure modes. Seven end-to-end real-world examples demonstrate the approach in practice. By retaining Excel as a familiar interface for input and output while integrating version control, automated refreshing, and seamless incorporation of statistical and machine learning methods, the proposed framework enables governed, reproducible, and auditable tabular data analysis.

0 citationsRead paper

Generating Query-Focused Summarization Datasets from Query-Free Summarization Datasets

May 06, 2026

This work addresses the scarcity of query-focused summarization (QFS) datasets by proposing a novel evidence-based method for automatically generating high-quality queries from plain document-summary pairs. By leveraging pretrained language models in conjunction with state-of-the-art QFS systems, the approach effectively bridges the gap between standard summarization data and query-focused settings. Both intrinsic and extrinsic evaluations demonstrate the method’s strong performance: summaries generated using the automatically produced queries achieve ROUGE scores comparable to those obtained with human-written queries, confirming the validity and practical utility of the proposed framework.

0 citationsRead paper

Text Summarization With Graph Attention Networks

Apr 04, 2026

This study investigates how to effectively integrate discourse structure—specifically Rhetorical Structure Theory (RST) and coreference graphs—to enhance abstractive text summarization. To this end, the authors present the first RST annotations for the XSum dataset, establishing a new benchmark named XSum-RST, and propose a graph fusion architecture combining Graph Attention Networks (GATs) and Multilayer Perceptrons (MLPs). Experimental results demonstrate that the MLP-based fusion approach outperforms baseline methods on the CNN/DailyMail dataset. Furthermore, evaluations on XSum-RST validate the effectiveness of incorporating graph-structured discourse information and delineate its practical limits, revealing notable differences in how discourse structures influence summary generation across distinct datasets.

0 citationsRead paper

Recent publications

Latest Papers

Graph-Based Agentic AI with LangGraph: Workflow Pathways for Long-Running Stateful Business Processes

Jul 21, 2026

This work addresses the challenges of constructing and managing generative AI agent systems for long-horizon, stateful, multi-step business processes by proposing a graph-structured workflow design methodology. Leveraging the LangGraph framework, it explicitly models core mechanisms such as state management, conditional routing, and human-in-the-loop interventions. The approach is instantiated in three representative applications: SQL analysis with repair loops, retrieval-augmented generation gated by evidential validation, and human-AI collaborative policy review supporting interruption and checkpoint-based recovery. By treating behaviors like routing, pausing, and audit trails as explicit product features rather than implicit prompt logic, this study not only delineates the applicability boundaries of LangGraph in high-complexity workflows but also substantially enhances system controllability, reliability, and auditability in real-world operational settings, establishing a reusable engineering paradigm.

0 citationsRead paper

From Patent Expiry to Business Pathways: AI Workflows for Activating Innovation Archives

Jul 11, 2026

This study addresses the underutilization of expired or lapsed patents, whose embedded technical knowledge often remains inaccessible due to difficulties in identification and interpretation, thereby hindering innovation transfer. The work proposes the first AI-driven framework that treats patent expiration as both a commercial opportunity and a knowledge archival milestone, integrating legal status, technical content, and market signals to generate structured business pathways—such as SaaS offerings or consulting services. The methodology combines patent metadata, fee payment records, semantic search, patent family analysis, and a locally deployed Qwen3.6 generative model to enable multidimensional risk assessment. Evaluated on the CIPO dataset, the approach successfully identified 20 high-potential patents, demonstrating the stability of the scoring model and compliance of AI-generated outputs, while also uncovering practical challenges such as insufficient legal status coverage.

0 citationsRead paper

Pandas for Reproducible Data Analysis: From Spreadsheets to Research-Grade Python Workflows

Jun 12, 2026

This work addresses the significant limitations of spreadsheet-based analysis in reproducibility, auditability, version control, and automation. It proposes a migration pathway from Excel to research-grade analytical workflows by leveraging Python’s pandas library as a bridge. The study introduces an innovative set of Excel-to-pandas mapping rules, categorizes nine canonical workflow patterns, and compiles a catalog of common failure modes. Seven end-to-end real-world examples demonstrate the approach in practice. By retaining Excel as a familiar interface for input and output while integrating version control, automated refreshing, and seamless incorporation of statistical and machine learning methods, the proposed framework enables governed, reproducible, and auditable tabular data analysis.

0 citationsRead paper

Generating Query-Focused Summarization Datasets from Query-Free Summarization Datasets

May 06, 2026

This work addresses the scarcity of query-focused summarization (QFS) datasets by proposing a novel evidence-based method for automatically generating high-quality queries from plain document-summary pairs. By leveraging pretrained language models in conjunction with state-of-the-art QFS systems, the approach effectively bridges the gap between standard summarization data and query-focused settings. Both intrinsic and extrinsic evaluations demonstrate the method’s strong performance: summaries generated using the automatically produced queries achieve ROUGE scores comparable to those obtained with human-written queries, confirming the validity and practical utility of the proposed framework.

0 citationsRead paper

Text Summarization With Graph Attention Networks

Apr 04, 2026

This study investigates how to effectively integrate discourse structure—specifically Rhetorical Structure Theory (RST) and coreference graphs—to enhance abstractive text summarization. To this end, the authors present the first RST annotations for the XSum dataset, establishing a new benchmark named XSum-RST, and propose a graph fusion architecture combining Graph Attention Networks (GATs) and Multilayer Perceptrons (MLPs). Experimental results demonstrate that the MLP-based fusion approach outperforms baseline methods on the CNN/DailyMail dataset. Furthermore, evaluations on XSum-RST validate the effectiveness of incorporating graph-structured discourse information and delineate its practical limits, revealing notable differences in how discourse structures influence summary generation across distinct datasets.

0 citationsRead paper