Reproduce “TSFLab: A Self-Improving Agent-Agnostic Workbench for Time Series Forecasting Research” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
ChengAo Shen, Yanjun Zhao, Haoxin Liu, Bonnie Chen, Miles Ma, Dongjin Song, Hanghang Tong, Qingsong Wen and 2 more
TSFLab provides shared infrastructure for time-series forecasting research, combining a domain guide, source-backed knowledge, reusable components, and a common experiment runner. In tests with Codex CLI, it completed 12/12 reproduction sessions versus TSLib’s 10/12, using 0.434 versus 0.533 million fresh tokens per accepted session. Across 96 scheduled reproductions, 73% were within 10% of paper MSE. These results concern tested tasks.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.