TSCenter/awesome-time-series-papers

What it solves

This project provides a centralized, curated repository of the latest research papers and code implementations in time series analysis. It helps researchers and practitioners quickly discover state-of-the-art methods across a broad spectrum of time series tasks, saving them the effort of manually searching through proceedings from major AI conferences.

How it works

The repository acts as an organized, community-maintained index. It categorizes papers by specific sub-fields—such as forecasting, anomaly detection, and foundation models—and tracks them by publication venue and year. Many entries include direct links to the original research papers and their associated code repositories, often highlighting highly cited or personally recommended works.

Who it’s for

This resource is designed for AI researchers, data scientists, and students who need to stay current with the rapidly evolving landscape of time series machine learning. It is particularly useful for those looking for benchmarks, implementation examples, or the latest theoretical advancements in the field.

Highlights

• Comprehensive coverage of major AI venues including NeurIPS, ICML, KDD, ICLR, and AAAI.
• Broad topical scope covering forecasting, anomaly detection, foundation models, causal discovery, and more.
• Active community participation with regular updates and support for pull requests.
• Visual indicators for high-citation papers and personal recommendations to help navigate the literature.