Nashville, TN & Williamsburg, VA – 24 Nov 2025 – A new study published in Artif. Intell. Auton. Syst. delivers the first systematic cross-model analysis of prompt engineering for structured data generation, offering actionable guidance for developers, data scientists, and organizations leveraging large language models (LLMs) in healthcare, e-commerce, and beyond. Led by Ashraf Elnashar from Vanderbilt University, alongside co-authors Jules White (Vanderbilt University) and Douglas C. Schmidt (William & Mary), the research benchmarks six prompt styles across three leading LLMs to solve a critical challenge: balancing accuracy, speed, and cost in structured data workflows.

image: Evaluates six prompt styles (JSON, CSV, Prefix, YAML, Function, Hybrid) across three LLMs (ChatGPT-4o, Claude, Gemini) on datasets (Stories, Medical, Receipts), measuring Accuracy, Token Cost, and Time to reveal model trade-offs and optimal prompt choices.

Credit: Ashraf Elnashar, Jules White/Vanderbilt University, Douglas C. Schmidt/William & Mary

Structured data—from medical records and receipts to business analytics—powers essential AI-driven tasks, but its quality and efficiency depend heavily on how prompts are designed. “Prior research only scratched the surface, testing a limited set of prompts on single models,” said Elnashar, the study’s corresponding author and a researcher in Vanderbilt’s Department of Computer Science. “Our work expands the horizon by evaluating six widely used prompt formats across ChatGPT-4o, Claude, and Gemini, revealing clear trade-offs that let practitioners tailor their approach to real-world needs.”

Key Findings: Accuracy vs. Efficiency—A Clear Choice for Every Use Case

The team’s rigorous experiment, conducted across three datasets (personal stories, medical records, and receipts), measured accuracy, token cost (a key driver of API expenses), and generation time for each prompt style-LLM combination. The results uncovered distinct strengths in each model:

• Claude emerged as the accuracy leader (85% overall), excelling with hierarchical prompt formats like JSON and YAML—ideal for complex, high-stakes tasks such as medical record generation where data integrity is non-negotiable.
• ChatGPT-4o stood out for efficiency, delivering the lowest token usage (under 100 tokens for lightweight formats) and fastest processing times (4–6 seconds on average), making it perfect for cost-sensitive or real-time applications like e-commerce receipt processing.
• Gemini offered a balanced middle ground, with solid performance across all metrics—though it showed variability with mixed-format prompts like Hybrid CSV/Prefix.

“Hierarchical formats like JSON and YAML boost accuracy but come with higher token costs, while lightweight options like CSV and simple prefixes cut latency without sacrificing much precision,” Elnashar explained. “For example, a healthcare provider handling patient data might prioritize Claude + JSON for accuracy, while an e-commerce platform could opt for ChatGPT-4o + CSV to process thousands of receipts efficiently.”

The study also highlighted a universal challenge: all LLMs struggled with narrative-style unstructured data (e.g., personal stories), with accuracy dropping to ~40% across prompt styles—underscoring the need for tailored approaches for different data types.

Practical Tools for Developers: Reusable Resources to Accelerate AI Workflows

Beyond insights, the research provides tangible value for the AI community. The team has made datasets, prompt templates, validation scripts, and design guidelines publicly available on GitHub (https://github.com/elnashara/EfficientStructuringMethods/tree/main), enabling reproducibility and immediate adoption.

“We wanted to move beyond theory—these resources let developers skip the trial-and-error and directly apply our findings to their pipelines,” said Jules White, co-author and professor at Vanderbilt’s Department of Computer Science. “Whether you’re building a medical data system or an e-commerce analytics tool, our work gives you a roadmap to choose the right prompt style and LLM.”

Looking Ahead: Expanding the Boundaries of Prompt Engineering

The study builds on the authors’ prior work focused on GPT-4o, now generalized to multiple models and prompt formats. Future research will explore LLMs’ robustness to noisy instructions, missing fields, and unseen schemas—critical considerations for real-world deployments. “As AI becomes more integrated into critical systems, we need to understand how these models perform when faced with the messiness of real data,” noted Schmidt, a professor in William & Mary’s Department of Computer Science.

This research was conducted without specific grant funding. The authors acknowledge the support of LLMs ChatGPT-4o, Claude, and Gemini for code generation, data visualization, and comparative evaluation.

About the Authors

• Ashraf Elnashar: Department of Computer Science, Vanderbilt University (ashraf.elnashar@vanderbilt.edu)
• Jules White: Department of Computer Science, Vanderbilt University
• Douglas C. Schmidt: Department of Computer Science, William & Mary

About the Publication

Title: Prompt engineering for structured data: a comparative evaluation of styles and LLM performanceJournal: Artif. Intell. Auton. Syst.DOI: 10.55092/aias2025009

License: Creative Commons Attribution 4.0 International License

Source from [https://www.eurekalert.org/news-releases/1107970].

Researchers have introduced a new blockchain-enabled framework that could significantly advance dynamic spectrum sharing (DSS) in future 6G wireless networks, addressing long-standing challenges in latency, security, and high-density spectrum coordination. Published in Blockchain, this work presents HierSpectrumChain, a hierarchical blockchain architecture that integrates smart-contract–driven Stackelberg auctions to coordinate spectrum access efficiently and securely among diverse wireless users.

image: Blockchain-enabled spectrum sharing architecture paving the way for secure, low-latency 6G networks.

Credit: Qin Wang/Nanjing University of Posts and Telecommunications, Muntasir Al Mamun/Nanjing University of Posts and Telecommunications.

Dynamic spectrum sharing is essential for next-generation systems, where billions of devices from autonomous vehicles to IoT sensors must coexist within limited frequency resources. Traditional centralized approaches struggle to deliver low latency, transparency, and trust across operators. To overcome these barriers, the research team led by Dr. Qin Wang at Nanjing University of Posts and Telecommunications, with contributors including Muntasir Al Mamun, developed a complete blockchain-based DSS system capable of real-time spectrum allocation at the network edge.

“HierSpectrumChain is designed to meet the stringent performance requirements of 6G,” explains Dr. Qin Wang. “By combining a main chain for global governance with localized sub-chains for edge-level transactions, the framework minimizes latency while maintaining a verifiable and secure global spectrum registry.” The system automates spectrum leasing through a two-stage Stackelberg auction encoded in smart contracts, allowing providers to publish available spectrum and secondary users to submit encrypted bids. All auction logic equilibrium computation, winner selection, and token settlement executes on-chain without requiring manual intervention.

A key innovation lies in pushing transaction processing to lightweight sub-chains, which reduces congestion on the main chain and supports high-frequency, micro-scale spectrum trades. The prototype implementation on a Ganache-based Ethereum test environment demonstrated that complete on-chain auction cycles can be executed in under 300 ms for up to 20 concurrent bidders illustrating the potential for real-time responsiveness in dense 6G scenarios. The experiments also reached throughput levels exceeding 80 transactions per second, with stable performance across repeated trials.

To ensure robust and scalable operation, the researchers conducted an in-depth analysis of consensus mechanisms, examining PoW, PoS, BFT, and DAG-based designs and highlighting how different models affect latency, fairness, and energy consumption. Building on this analysis, the team proposes an adaptive sub-chain consensus strategy capable of switching between PoS during low-traffic periods and BFT during peak demand, helping the system maintain predictable confirmation times while keeping energy usage low.

“Blockchain has long promised decentralized trust, but real-time wireless environments require extreme efficiency,” notes the research team. “Our hierarchical design demonstrates how both goals can be achieved simultaneously.” Beyond the architecture itself, the study also discusses practical considerations such as cross-chain interoperability, privacy preservation through encrypted bidding, and regulatory compliance for decentralized spectrum transactions.

Looking ahead, the team plans to extend HierSpectrumChain to permissioned platforms such as Hyperledger Fabric and FISCO-BCOS, conduct multi-peer experiments under realistic network latency, and integrate zero-knowledge proofs to further strengthen privacy in competitive bidding scenarios. The authors believe that this research could lay the groundwork for secure and transparent spectrum markets, enabling more efficient wireless communication and smoother coexistence among operators in the 6G era.

This paper, “Blockchain-Enabled Dynamic Credible Spectrum Sharing in 6G Networks,” was published in Blockchain.

Wang Q, Mamun M.A., et al. Blockchain-Enabled Dynamic Credible Spectrum Sharing in 6G Networks. Blockchain 2025. https://doi.org/10.55092/blockchain20250014

Source from [https://www.eurekalert.org/news-releases/1108275].

12月30日，昆明理工大学官网发布消息：日前，经云南省委同意：云南省政府聘任杨斌同志为昆明理工大学校长（聘期五年）。

此番履新前，他是昆明理工大学副校长。他本科毕业于东北大学，硕博均毕业于昆明理工大学冶金系，从事有色金属冶金的科研教学工作30余年。此次履新，也标志着杨斌院士正式掌舵母校。

杨斌简介

杨斌，男，汉族，1965年5月生，中国工程院院士，有色金属冶金专家，有色金属真空冶金创新团队首席科学家。

杨斌于1987年获东北工学院学士学位；1990年获昆明工学院（现昆明理工大学）冶金系硕士学位，同年起留校任教；1998年获昆明理工大学博士学位；2007年1月至12月赴莫斯科国立钢铁及合金学院访问学习。历任昆明理工大学材料与冶金工程学院副院长，昆明理工大学科技处处长兼省知识产权发展研究院院长等职；2017年起任昆明理工大学副校长；2021年起出任真空冶金国家工程研究中心主任；2025年11月21日当选中国工程院院士。2025年12月担任昆明理工大学校长。

杨斌主要从事有色金属冶金的科研教学工作，主持国家高技术产业化示范项目、国家自然科学基金重点项目、国家重点研发计划课题等科研项目40项，发表SCI、EI收录论文550篇，出版学术著作5部，获授权发明专利280件，组织制定国家标准1项。先后获国家技术发明奖二等奖2项（排名一、三），国家科技进步奖二等奖2项（均排名一），何梁何利基金科技创新奖1项，省部级科技一等奖8项（均排名一），中国专利优秀奖1项，中国产学研合作创新成果奖2项，成果两次入选云南十大科技进展。

来源：云南日报-云新闻、昆明理工大学，仅用于学术分享，如有侵权请联系删除。

据“微言教育”微信公众号消息，12月30日，教育部人事司在华东师范大学宣布了教育部党组的任免决定，马余刚同志任华东师范大学校长、党委副书记，钱旭红同志不再担任华东师范大学校长、党委副书记职务。教育部人事司主要负责同志出席会议并讲话。上海市教卫工作党委主要负责同志出席会议。

马余刚简介

马余刚，男，汉族，1968年3月生，中共党员，研究生学历，博士学位，教授、博士生导师，中国科学院院士。

马余刚于1989年从杭州大学物理学系毕业；1994年获得中国科学院上海原子核研究所博士学位，之后留所工作；1997年获得国家杰出青年科学基金资助；1999年入选“百千万人才工程”国家级第一、二层次人选；2014年担任中国科学院上海应用物理研究所副所长；2015年当选为美国物理学会会士；2017年当选为中国科学院院士；2023年任复旦大学副校长；2024年兼任复旦大学研究生院院长；2025年12月，任华东师范大学校长、党委副书记。

马余刚主要从事重离子核物理实验与唯象研究，在中高能及极端相对论的重离子碰撞实验和理论研究等前沿领域取得多项重要成果，其中有关反物质研究的成果两次入选“中国科学十大进展”。曾获国家自然科学二等奖（2016年度、2001年度）、全球华人物理与天文学会-亚洲成就奖（2015年度）、上海市自然科学奖一等奖（2014年度）、中国青年科技奖等多项荣誉。

来源：微言教育、复旦大学、华东师范大学，仅用于学术分享，如有侵权请联系删除。