Reproduce “CASM: Context-Aware Semi-Markov Post-Processor for Beat Tracking” 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
Zhanhong He, Hanyu Meng, Yaolong Ju
CASM is an offline, model-agnostic decoder that turns beat activations into event times by balancing local evidence against rhythmic regularity. It estimates local periods and measures whether the best period is better than alternatives, relaxing regularization when timing is ambiguous. On fixed activations, it improves every reported metric for three backbones on GTZAN and SMC, after calibration on SMC folds 1–7.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.