Submission Instructions

We welcome full paper submissions. The papers must be no longer than 8 pages in total (excluding references). Please submit at the following CMT website:
ADFM 2025 CMT submission website.
The Microsoft CMT service was used for managing the peer-reviewing process for this conference. This service was provided for free by Microsoft and they bore all expenses, including costs for Azure cloud services as well as for software development and support.

• All submissions are handled via the workshop’s CMT website.
• Submissions should be made in PDF format and should follow the official ICCV 2025 template and guidelines.
• All submissions must be anonymous and conform to the ICCV 2025 conference guidelines for double-blind review.
• ADFM 2025 workshop follows the ICCV 2025 conference guidelines and does not allow dual submissions.
• Authors may upload optional supplementary materials, containing additional details, videos, images, etc. in a separate zip file (with a max of 50MB in size); the deadline for submitting these supplementary materials is the same as that for the main paper.
• The accepted papers will be presented as either an oral, spotlight, or poster presentation.
• By submitting a paper to the ADFM 2025 workshop for review, the authors must agree that at least one author or authors' representative of each accepted submission must present the paper at the workshop in-person and that they are willing and able to serve as the reviewers of the ADFM 2025 workshop submissions if needed (decided by the ADFM 2025 workshop organizing team).
• The presentation of accepted papers at our workshop will follow the same policy as that for the accepted papers at the ICCV 2025 main conference.
• The accepted papers will be made publicly accessible on the workshop website shortly after the camera-ready deadline. ICCV 2025 will provide the official proceedings of the accepted papers.
• Failure to comply with the aforesaid rules may cause the paper to be removed from the workshop program.

Submission deadline:       June 20 30, 2025 11:59 PM EDT (extended)  
Notification to authors:    July 11, 2025
Camera ready deadline:  August 18, 2025

We invite the submission of original and high-quality research papers in the topics related to anomaly detection with foundation models.

We're seeking dedicated Reviewers! Please self-nominate via the reviewer self-nomination form. Thanks for your support!

Topics

The topics for ADFM 2025 include, but are not limited to:

• Fundamental theories and principles of foundation models for anomaly detection.
• Advanced anomaly detection algorithms and frameworks utilizing foundation models.
• Sector-specific anomaly detection employing foundation models, covering areas such as finance, healthcare, cybersecurity, and industrial systems.
• Evaluation standards and benchmarks for appraising anomaly detection in foundation models.
• Methods enhancing the clarity and comprehensibility of foundation models in anomaly detection.
• Methods promoting fairness and diminishing bias in anomaly detection with foundation models.
• Privacy-enhancing techniques in anomaly detection with foundation models.
• Trust and reliability of foundation models in crucial anomaly detection applications.
• Interdisciplinary methods for refining anomaly detection, incorporating insights from fields like psychology and sociology.
• Adaptive learning and adjustment mechanisms for foundation models in dynamic settings.
• Integration of expert knowledge and domain-specific systems with foundation models for enhanced anomaly detection.
• Exploratory discussions on the constraints and challenges of current foundation models in identifying anomalies in complex and noisy datasets.
• Prospective insights on the evolution of anomaly detection methods with the advancement of foundation models.

 

• A Comprehensive Library for Benchmarking Multi-class Visual Anomaly Detection.
  Zhang, Jiangning; He, Haoyang; Gan, Zhenye; He, Qingdong; Cai, Yuxuan; Xue, Zhucun; Wang, Yabiao; Wang, Chengjie; Xie, Lei; Liu, Yong.


• Bridge Feature Matching and Cross-Modal Alignment with Mutual-filtering for Zero-shot Anomaly Detection.
  Bai, Yuhu; Zhang, Jiangning; Cao, Yunkang; Lu, Guangyuan; He, Qingdong; Li, Liangtai; Tian, Guanzhong.


• DinoAtten3D: Slice-Level Attention Aggregation of DinoV2 for 3D Brain MRI Anomaly Classification.
  Rafsani, Fazle; Shah, Jay; Chong, Catherine D.; Schwedt, Todd J.; Wu, Teresa.


• Zero-Shot Image Anomaly Detection Using Generative Foundation Models.
  Abdi, Lemar; Valiuddin, Amaan; Caetano, Francisco; Viviers, Christiaan; van der Sommen, Fons.

Aishwarya Budhkar	Indiana University
Anshuman Kumar	University of California Davis
Ashish Singh	University of Massachusetts Amherst
Bingke Zhu	Institute of Automation, Chinese Academy of Sciences
Haoyang He	Zhejiang University
Huimin Xie	TikTok Inc.
Manyi Yao	University of California, Riverside
Niv Cohen	New York University
Qingdong He	Tencent
Rakshith Mahishi	University of California, Riverside
Shravya Kanchi	Virginia Tech
Yunkang Cao	Hunan University
Yuxuan Cai	Huazhong University of Science and Technology
Zhuang Luo	WPI