Please note, this is the web page of the 2025 LLM4Sec Workshop edition, provided for archival purposes. For the most current edition, please click here.

logo

November 13, 2025 - Washington DC, USA

alt text

The LLM4Sec workshop will be co-located with the
IEEE International Conference on Data Mining (ICDM) 2025

Sponsored by

alt text

Call For Papers

Important Dates

Paper Submission Deadline August 29, 2025
Notification to authors September 15, 2025 extended: September 18, 2025
Camera-ready versions September 25, 2025 extended: October 5, 2025
Workshop date November 13, 2025

Note: all deadlines are Anywhere On Earth (AOE) time

Submission site: https://ieeecps.org/cps/v2/auth/login?ak=1&pid=24XmCKZ1gAf1u24iyCTGnr

CFP

Large Language Models (LLMs) are widely used for their exceptional ability in performing natural language processing applications like question answering, text completion, and text translation, to name a few. These capabilities enable their use in several domains such as customer support and interaction, content creation, editing and proofreading, sentiment analysis, etc. Besides the natural language, LLMs can generate and manipulate sequences of tokens of any kind, acting as boxes into which human knowledge can be compressed and then extracted when necessary. Owing to this, LLMs can be used to solve a wide range of problems and have been increasingly incorporated into several software frameworks. Among the others, their adoption to advance in the field of cyber security is gaining momentum. As a matter of fact, LLMs have been employed to expose and remediate security flaws, generate secure code and test cases, detect vulnerable or malicious code, and verify the integrity, confidentiality, and reliability of data. Interesting results have been presented so far, but the research in this area is still in its early stages, and it has the potential to produce further significant findings.

This workshop aims to stimulate research on LLM-based solutions for security and privacy. We invite both academic and industrial researchers to submit research papers as either original works, discussion papers, or excerpt of published articles.

Topics of interest include, but are not limited to:

  • Secure code generation
  • Test case generation
  • Vulnerable code detection
  • Malicious code detection
  • Vulnerable code fixing
  • Software deobfuscation and repairing
  • Anomaly-based detection
  • Signature-based detection
  • Network security
  • Computer forensics
  • Spam detection
  • Phishing detection and prevention
  • Vulnerability discovery
  • Malware identification and analysis
  • Data anonymization/de-anonymization
  • Big data analytics for security
  • Data integrity
  • Data confidentiality
  • Data reliability
  • Data traceability
  • Zero-day attack detection
  • Automated security policy generation
  • Predictive analytics
  • Decision support

Submission guidelines

  • Authors are invited to submit original papers that have not been published elsewhere and are not currently under consideration for another journal, conference, or workshop.
  • Submissions must be a PDF file limited to a maximum of eight (8) pages (plus 2 extra pages for bibliography and any appendices) in the IEEE 2-column format (https://www.ieee.org/conferences/publishing/templates.html), including the bibliography and any appendices. Submissions longer than 10 pages will be rejected without review. All submissions will be triple-blind reviewed by the Program Committee based on technical quality, relevance to the scope of the workshop, originality, significance, and clarity.
  • Manuscripts must be submitted electronically through the online submission system: https://ieeecps.org/cps/v2/auth/login?ak=1&pid=24XmCKZ1gAf1u24iyCTGnr
  • Accepted papers will be included in the ICDM Workshop (ICDMW) proceedings (separate from ICDM Main Conference Proceedings), published by the IEEE Computer Society Press in IEEExplore and indexed in Scopus.
  • At least one author of each accepted paper must complete the registration and present the paper at the workshop for it to be included in the proceedings and program. Registration link and fees can be found at https://www3.cs.stonybrook.edu/~icdm2025/registration.html

Program

Accepted Papers (ICDMW Proceedings)

  • Zefang Liu, Arman Anwar - AutoBnB-RAG: Enhancing Multi-Agent Incident Response with Retrieval-Augmented Generation
  • Christopher Beddies, Vincenzo Bonaiuto, Bernd Eylert, Stefan Kubica, Bernd Heimer - Deterministic Network Vulnerability Test Generation Through Predictable Template Learning: A Systematic Evaluation
  • Lauren Brandt, Joshua Guttman, Andres Molina-Markham - Autoformalization of Cryptographic Protocols
  • Steve Huntsman - Coherence-Driven Inference for Cybersecurity
  • Giacomo Benedetti, Luca Caviglione, Michał Choraś, Massimo Guarascio, Angelica Liguori, Giuseppe Manco, Antonino Rullo - Days of Future Past: Towards Robust Detection of Malware Variants via LLM-Based Embedding Generation
  • Shota Sakamaki, Hiroki Kanezashi, Toyotaro Suzumura, and Tomohiro Kudoh - Enhancing Graph-Based Advanced Persistent Threat Detection with LLM Sentence Embedding
  • Philippe Zotz, Beltran Fiz - MISPerer: Evaluating LLM-Mediated Access to MISP via the Model Context Protocol
  • Han Wang, Murathan Kurfalı, Alfonso Iacovazzi - Benchmarking LLMs for Threat Level Determination
  • Aidan Wen, Norah A. Alzahrani, Jingzhi Jiang, Andrew Joe, Karen Shieh, Andy Zhang, Basel Alomair, David Wagner - SeedAIchemy: LLM-Driven Seed Corpus Generation for Fuzzing
  • Yu-Zheng Lin, Sujan Ghimire, Abhiram Nandimandalam, Jonah Michael Camacho, Unnati Tripathi, Rony Macwan, Sicong Shao, Setareh Rafatirad, Rozhin Yasaei, Pratik Satam, Soheil Salehi - LLM-HyPZ: Hardware Vulnerability Discovery Using an LLM-Assisted Hybrid Platform for Zero-Shot Knowledge Extraction and Refinement
  • Mudita Khurana, Raunak Jain - SoK: Measuring What Matters for Closed-Loop Security Agents
  • Hamed Jelodar, Mohammad Meymani, Samita Bai, Roozbeh Razavi-Far, Ali A. Ghorbani - SBAN: A Framework & Multi-Dimensional Dataset for Large Language Model Pre-Training and Software Code Mining

Workshop Program

10:30 am Welcome Message and Opening Remarks (Antonino Rullo)
Session 1 - Autonomous and Multi-agent Systems for Cyber Defense (session chair Antonino Rullo)
10:40 am AutoBnB-RAG: Enhancing Multi-Agent Incident Response with Retrieval-Augmented Generation - Zefang Liu, Arman Anwar
11:05 am Coherence-driven inference for cybersecurity - Steve Huntsman
11:30 am SoK: Measuring What Matters for Closed-Loop Security Agents - Raunak Jain, Mudita Khurana
12:00 am Lunch break
Session 2 - Generative and Analytical AI for Vulnerability Discovery (session chair Zefang Liu)
1:30 pm Deterministic Network Vulnerability Test Generation through Predictable Template Learning: A Systematic Evaluation - Christopher Beddies, Vincenzo Bonaiuto, Bernd Eylert, Stefan Kubica, Bernd Heimer
1:55 pm LLM-HyPZ: Hardware Vulnerability Discovery using an LLM-Assisted Hybrid Platform for Zero-Shot Knowledge Extraction and Refinement (short paper) - Yu-Zheng Lin, Sujan Ghimire, Abhiram Nandimandalam, Jonah Michael Camacho, Veeramani Pugazhenthi, Unnati Tripathi, Rony Macwan, Sicong Shao, Setareh Rafatirad, Rozhin Yasaei, Pratik Satam, Soheil Salehi
2:10 pm SeedAIchemy: LLM-driven Seed Corpus Generation for Fuzzing - Aidan Wen, Norah A. Alzahrani, Jingzhi Jiang, Andrew Joe, Karen Shieh, Andy Zhang, David Wagner, Basel Alomair
2:35 pm Short break
Session 3 - Advanced Malware and Attack Detection through LLMs (session chair Antonino Rullo)
2:45 pm Days of Future Past: Towards Robust Detection of Malware Variants via LLM-based Embedding Generation - Giacomo Benedetti, Luca Caviglione, Michał Choraś, Massimo Guarascio, Angelica Liguori, Giuseppe Manco, Antonino Rullo
3:10 pm Enhancing Graph-based Advanced Persistent Threat Detection with LLM Sentence Embedding (short paper) - Shota Sakamaki, Hiroki Kanezashi, Toyotaro Suzumura, Tomohiro Kudoh
3:30 pm SBAN: A Multi-Dimensional Dataset for Pre-Training LLMs and Software Code Mining - Hamed Jelodar, Mohammad Meymani, Samita Bai, Roozbeh Razavi-Far
4:00 pm Coffee break
Session 4 - LLM-Augmented Intelligence, Formal Reasoning, and Secure Communication (session chair Steve Huntsman)
4:30 pm Autoformalization of Cryptographic Protocols - Lauren Brandt, Joshua Guttman, Andres Molina-Markham
4:55 pm Benchmarking LLMs for Threat Level Determination - Han Wang, Murathan Kurfalı, Alfonso Iacovazzi
5:20 pm MISPerer: Democratizing Threat Intelligence using Large Language Models (short paper) - Philippe Zotz, Beltran Fiz
5:35 pm Closing remarks (Antonino Rullo)

Organizers

Chairs

Program Committee

  • Annalisa Appice, University of Bari Aldo Moro, Italy
  • Akshata Bhat, Amazon AGI, USA
  • Giacomo Benedetti, CNR – IMATI, Italy
  • Carlos Catania, Universidad Nacional de Cuyo, Argentina
  • Tianyu Chen, Microsoft Research Asia
  • Daiki Chiba, NTT Security Holdings Corporation, Japan
  • Jürgen Cito, TU Wien, Austria
  • Moumita Das Purba, University of North Carolina, USA
  • Zhaopeng Gu, Chinese Academy of Sciences, China
  • Venkata Gopi, Salesforce, USA
  • Andreas Happe, Technical University of Vienna, Austria
  • Michele Ianni, Università della Calabria, Italy
  • Zhengjie Ji, Virginia Tech, USA
  • Nan Jiang, Purdue University, USA
  • Mudita Khurana, Airbnb, USA
  • Mohammed Latif Siddiq, University of Notre Dame, France
  • Yeting Li, Chinese Academy of Sciences, China
  • Puzhuo Liu, Tsinghua University, China
  • Thibaud Lutellier, University of Alberta, Canada
  • Wojciech Mazurczyk, Warsaw University of Technology, Poland
  • Surendra Narang, Palo Alto Networks, USA
  • Antonino Nocera, University of Pavia, Italy
  • Sudipta Paria, University of Florida, USA
  • Jiaxing Qi, Beihang University, China
  • Rajat Sadhukhan, Indian Institute of Roorkee, India
  • P.V. Sai Charan, New York University, USA
  • Gustavo Sandoval, New York University, USA
  • Edoardo Serra, Boise State University, USA
  • Francesca Spezzano, Boise State University, USA
  • Veronica Valeros, Czech Technical University, Czech Republic
  • Steffen Wendzel, University of Ulm, Germany
  • Ferhat Yaman, AMD, USA
  • Chenyuan Yang, University of Illinois Urbana-Champaign, USA
  • Jianxing Yu, Sun Yat-sen University, China
  • Cen Zhang, Georgia Institute of Technology, USA
  • Ying Zhang, Wake Forest University, USA

Gallery

alt text