三类AI挑战夺冠方案，涵盖图像检索、异常检测与模型安全。
No Trust Issues Here: A Technical Report on the Winning Solutions for the Rayan AI Contest
- 融合图文输入的图像检索系统，准确率达95.38%。
- 零样本异常检测模型在无异常训练下实现73.14%准确率。
- 可检测神经网络中的隐藏后门，准确率78%，适合安全领域应用。
本报告介绍在Rayan AI竞赛中解决的三项机器学习挑战：组合图像检索、零样本异常检测和后门模型检测。在组合图像检索任务中，我们开发了一个处理视觉与文本输入的系统，实现95.38%的准确率，并以明显优势排名第一。在零样本异常检测中，设计的模型可在未见异常样本情况下识别并定位图像异常，获得73.14%的得分，排名第二。在后门模型检测任务中，提出一种检测神经网络中隐藏后门触发器的方法，达到78%的准确率，位列第二。这些结果证明了方法在检索、异常检测与模型安全方面的有效性，对医疗、制造及网络安全等实际应用具有重要意义。所有解决方案代码已公开（https://github.com/safinal/rayan-ai-contest-solutions）。
原文摘要 · Abstract (English)
This report presents solutions to three machine learning challenges developed as part of the Rayan AI Contest: compositional image retrieval, zero-shot anomaly detection, and backdoored model detection. In compositional image retrieval, we developed a system that processes visual and textual inputs to retrieve relevant images, achieving 95.38% accuracy and ranking first with a clear margin over the second team. For zero-shot anomaly detection, we designed a model that identifies and localizes anomalies in images without prior exposure to abnormal examples, securing second place with a 73.14% score. In the backdoored model detection task, we proposed a method to detect hidden backdoor triggers in neural networks, reaching an accuracy of 78%, which placed our approach in second place. These results demonstrate the effectiveness of our methods in addressing key challenges related to retrieval, anomaly detection, and model security, with implications for real-world applications in industries such as healthcare, manufacturing, and cybersecurity. Code for all solutions is available online (https://github.com/safinal/rayan-ai-contest-solutions).
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。