DengBoCong/nlp-paper

Provides curated paper implementations with dual TensorFlow/PyTorch codebase across 30+ NLP subdomains including dialogue systems, text similarity, and knowledge graphs, with an integrated CLI search tool for paper discovery. Covers both classical algorithms (PageRank, K-Means) and modern architectures (pre-training, prompt learning, GNNs), organized chronologically with detailed Chinese reading notes and reproduced model code. Extends into specialized applications like task-oriented dialogue state tracking, relation extraction, and text summarization using graph-based and neural approaches.

1,331 stars. No commits in the last 6 months.