Uncertainty Decomposition via Cyclical SG-MCMC and Soft-label Learning for Subjective NLP 事件

Uncertainty Decomposition via Cyclical SG-MCMC and Soft-label Learning for Subjective NLP arXiv:2605.24773v1 Announce Type: new Abstract: Annotator disagreement in emotion classification reflects ambiguity intrinsic to emotion concepts and is essential for predictor-quality assessment in subjective NLP. Yet no prior work integrates soft-label learning with Bayesian deep learning to evaluate uncertainty along axes including annotator-distribution fidelity. We train a linear head on a frozen RoBE