Regime-based evaluation framework for financial NLP stability. Implements chronological cross-validation, semantic drift quantification via Jensen-Shannon divergence, and multi-faceted robustness profiling. Replicates Sun et al.'s (2025) methodology with modular, auditable Python codebase.

Quantifying lexical semantic change of 'toxic', 'healing', 'burnout' on Twitter via temporal Word2Vec + Procrustes alignment | displacement scores, nearest-neighbor shift, t-SNE visualization

Comparative corpus analysis for Python: keyness, collocations, semantic shift, temporal trajectories with changepoints + causal inference.