Reproduce “Semi-Supervised Data Programming with Subset Selection” 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
Ayush Maheshwari, Oishik Chatterjee, KrishnaTeja Killamsetty, Ganesh Ramakrishnan, Rishabh Iyer
SPEAR combines a feature-based classifier with a probabilistic label model for heuristic labeling functions, training them jointly with a small set of true labels and many unlabeled examples. SPEAR-SS also selects examples for labeling by entropy filtering and feature-space diversity. On YouTube spam, supervised selection gained 4.2 accuracy points over Only-L’s 90.7, but results varied by task.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.