// the find
yzhao062/anomaly-detection-resources
Anomaly detection related books, papers, videos, and toolboxes. Last update late 2025 for LLM and VLM works!
A well-maintained 'awesome list' for anomaly detection, covering papers, toolkits, datasets, courses, and benchmarks. Maintained by Yuxuan Zhao, the author of PyOD and several other tools in the space, so the selections reflect genuine domain expertise rather than link farming. Aimed at ML practitioners and researchers who need a starting point for outlier detection work.
The author has skin in the game — PyOD, SUOD, ADBench, and PyGOD are all their own work and genuinely useful libraries, not just citations to impress. The LLM/VLM section (4.1) was updated in late 2025 with ACL and NeurIPS papers, so it's not stale in the areas where things move fast. The toolbox section is organized by data type (multivariate, time series, graph) rather than just dumping every library in one pile, which makes it actually navigable. Benchmarks are called out separately with links to both paper and code, which is the right move — benchmarks without code are useless.
It's a link list, not a guide — no comparative notes, no 'start here if you're doing X' framing. Someone new to the field gets 200 links with no signal on what to actually read first. The papers section is heavily weighted toward the author's own lab (USC FORTIS) in the emerging topics and LLM sections, which is a real bias worth noting. The directory tree is essentially empty (README + two Python scripts), so there's no tooling, notebooks, or structured content beyond the RST document itself. Some subsections like 4.2 (Emerging Topics) have only 4 papers, which feels underpopulated compared to the rest.