Abstract: We consider the problem of zero-shot anomaly detection in which a model is pre-trained to detect anomalies in images belonging to seen classes, and expected to detect anomalies from unseen ...

We propose HtmlRAG, which uses HTML instead of plain text as the format of external knowledge in RAG systems. To tackle the long context brought by HTML, we propose Lossless HTML Cleaning and Two-Step ...

What if you could transform hours of audio into precise, actionable text with just a few lines of code? In 2025, this is no longer a futuristic dream but a reality powered by innovative speech-to-text ...