AI RESEARCH
STaR-DRO: Stateful Tsallis Reweighting for Group-Robust Structured Prediction
arXiv CS.AI
•
ArXi:2604.09737v1 Announce Type: cross Structured prediction requires models to generate ontology-constrained labels, grounded evidence, and valid structure under ambiguity, label skew, and heterogeneous group difficulty. We present a two-part framework for controllable inference and robust fine-tuning. First, we