2025年CCF形式化方法专委会青年学术论坛——大模型辅助的程序分析与验证专题研讨会


2025年7月12日(星期六)

杭州市西湖区余杭塘路866号浙江大学紫金港校区海纳苑4幢(理学部大楼)101会议室


此次研讨会聚焦大模型技术与程序分析和验证领域的深度结合,围绕“大模型时代下程序分析与验证的理论与实践”展开前沿探讨。论坛旨在探索大模型技术为形式化方法带来的新范式与新工具,推动程序自动化分析、智能验证与系统安全保障的技术突破。此次研讨会的成功举办将为该领域的研究者和从业者提供一个交流合作的平台,激发创新思维,推动形式化方法与AI结合的技术创新和应用落地,为行业发展提供新的思路和方法。

主办单位:中国计算机学会

承办单位:中国计算机学会形式化方法专委会、浙江大学

组织委员会

  • 吴志林,中国科学院软件研究所
  • 沈博,西北工业大学
  • 陈明帅,浙江大学
  • 王竟亦,浙江大学
  • 姚培森,浙江大学

腾讯会议链接https://meeting.tencent.com/dm/HXC1y2dy00if

直播链接http://live.bilibili.com/24106724

会议日程

报告人 报告主题 时间
论坛开幕 09:00-09:15
张业迪(新加坡国立大学) Towards Trustworthy AI Agents: Integrating Large Language Models with Formal Methods 09:15-10:00
茶歇 10:00-10:30
时清凯(南京大学) 基于分治的可靠大语言模型应用 10:30-11:15
李屹(华为操作系统部OS内核实验室) 大模型驱动的泛形式化软件工程实践 11:15-12:00
午休 12:00-14:00
刘尚清(南京大学) 基于大语言模型的程序规约理解与生成 14:00-14:45
曹嘉伦(香港科技大学) From Informal to Formal – Incorporating and Evaluating LLMs on Natural Language Description to Verifiable Formal Proofs 14:45-15:30
茶歇 15:30-16:00
李昊坤(华为2012可信费马实验室) SMT(NRA)高效求解:结合局部搜索,MCSAT和openCAD的混合方法 16:00-16:45
论坛闭幕 16:45-17:00
Towards Trustworthy AI Agents: Integrating Large Language Models with Formal Methods

photo

报告摘要: Large Language Models (LLMs) have emerged as a transformative AI paradigm, profoundly influencing broad aspects of daily life. Despite their remarkable performance, LLMs exhibit a fundamental limitation: hallucination—the tendency to produce misleading outputs that appear plausible. This inherent unreliability poses significant risks, particularly in high-stakes domains where trustworthiness is essential. On the other hand, Formal Methods (FMs), which share foundations with symbolic AI, provide mathematically rigorous techniques for modeling, specifying, reasoning, and verifying the correctness of systems. These methods have been widely employed in mission-critical domains such as aerospace, defense, and cybersecurity. However, the broader adoption of FMs remains constrained by significant challenges, including steep learning curves, limited scalability, and difficulties in adapting to the dynamic requirements of daily applications. To build trustworthy AI agents, we argue that the integration of LLMs and FMs is necessary to overcome the limitations of both paradigms. LLMs offer adaptability and human-like reasoning but lack formal guarantees of correctness and reliability. FMs provide rigor but need enhanced accessibility and automation to support broader adoption from LLMs.

报告人简介: Yedi Zhang obtained her Ph.D. in 2023 from ShanghaiTech Univeristy, where her research focused on the formal verification of neural networks and received the CCF outstanding doctoral dissertation award in formal methods. After that, she joined National University of Singapore, Singapore, as a postdoc to continue on verification and validation of AI systems. Her current research explores the integration of formal methods with large language methods to advance the development of trustworhty AI agents. She serves and has served as a program committee member or reviewer for various conferences/journals (e.g., CAV, FAC, ATVA, TASE, ICSE, TOSEM, APSEC, ICLR, NeurIPS, ACM MM) in the fields of formal methods, software engineering, and artificial intelligence. More infomration at: https://zhangyedi.github.io/.

基于分治的可靠大语言模型应用

img

报告摘要: 当前,大语言模型正以各种形式重塑包括形式化方法在内的各种技术领域,为提高软件开发效率、保障软件质量提供了新思路、新方案。然而,大语言模型不是“银弹”,其存在诸如幻觉、符号限制、缺乏领域知识等问题,让软件开发者在很多情况下不能通过简单的问答获得满意的结果。因此,如何在实践中可靠地使用大语言模型,克服上述诸多问题,成为一项亟需解决的难题。该报告将探讨大语言模型应用的一些想法,提出对“可靠使用大语言模型”的两个思路 —— 通过分治算法缓解大语言模型幻觉及符号限制、通过可靠的结果反馈增强大语言模型领域知识 —— 以及这两个思路在网络协议安全分析中的应用。

报告人简介: 时清凯,南京大学计算机学院副教授,博士生导师,国家级青年人才。2020年于香港科技大学获得博士学位,曾任源伞科技联合创始人、蚂蚁集团技术专家、美国普渡大学博士后研究员,目前主要从事编译技术及基于编译的软件安全技术研究,其研究成果广泛发表于诸如PLDI、OOPSLA、ICSE、FSE、SP、CCS等程序设计语言、软件工程、网络安全领域的CCF A类会议或期刊,其中一作论文10篇,并受邀担任相关会议审稿人,曾获ACM SIGPLAN或SIGSOFT杰出论文奖4次、Google论文奖、Hong Kong PhD Fellowship,曾三次参加全国软件大会原型竞赛均获一等奖。

From Informal to Formal – Incorporating and Evaluating LLMs on Natural Language Description to Verifiable Formal Proofs

photo

报告摘要: The research in AI-based formal mathematical reasoning has shown an unstoppable growth trend. These studies have excelled in mathematical competitions like IMO and have made significant progress. This paper focuses on formal verification, an immediate application scenario of formal reasoning, and breaks it down into sub-tasks. We constructed 18k high-quality instruction-response pairs across five formal specification languages (Coq, Lean4, Dafny, ACSL, and TLA+) by distilling gpt-4o and evaluated against ten open-sourced LLMs, including recent popular DeepSeek-R1. We also fine-tuned several 7~8B small models to achieve comparable performance with Deepseek-R1-671B. Interestingly, we observed that fine-tuning with formal data also enhances mathematics, reasoning, and coding capabilities.

报告人简介: Dr. Jialun Cao is a Research Assistant Professor at the Hong Kong University of Science and Technology. Her main research areas include AI testing, AI4SE, and AI4formal verification, etc. She received her Ph.D. from the Hong Kong University of Science and Technology under the supervision of Professor Shing-Chi Cheung. She has published more than 20 papers in ICSE, ESEC/FSE, CAV, ASE, USENIX Security, TOSEM. She serves as a program committee member of top-tier conferences including ICSE, FSE and ASE. She received the 2025 ACM SIGSOFT Outstanding Dissertation Award.

SMT(NRA)高效求解:结合局部搜索,MCSAT和openCAD的混合方法

photo

报告摘要: 本报告围绕非线性实数算术的可满足性问题(SMT-NRA),介绍我们在多项式公式求解与局部搜索方法方面的最新研究进展。我们首先提出了一种面向仅包含严格不等式的 SMT-NRA 子类的局部搜索算法。该算法基于实根隔离,设计了一种名为 Cell-Jump 的新型跳跃操作,可沿指定方向跨越由多项式集合在实数空间中划分的胞腔(cell),从而实现原生支持非线性实数算术的局部搜索策略。 在此基础上,我们进一步提出了 二维 Cell-Jump(2D Cell-Jump) 操作,增强搜索的空间跳跃能力,并据此构建了局部搜索框架 2d-LS。为提升求解效率,我们将该框架与模型构造型可满足性演算(MCSAT)相结合,并引入 Sample-Cell 投影算子,用于在实数域中引导搜索避开冲突状态。最终,我们设计了一个融合 2d-LS、MCSAT 与 OpenCAD 的混合求解框架,通过信息交互进一步提升整体性能。实验结果表明,该方法在搜索效率和求解能力方面均取得了显著提升。

报告人简介: 李昊坤,2022年获北京大学博士学位,现于华为2012可信费马实验室形式化方法专家,参与多个重要的形式化验证项目。主要研究方向为符号计算,定理证明及程序验证,相关工作发表在CAV,ACL, ISSAC,JSC等国际知名期刊和会议上。