A curated list of 100 foundational and influential papers in natural language processing for students and researchers.
mhagiwara/100-nlp-papers is a curated repository of 100 must-read academic papers and educational resources in natural language processing (NLP). It serves as a structured guide to the field's foundational concepts, breakthrough research, and modern developments, compiled to help students and researchers efficiently navigate the essential literature. The list is community-maintained and includes diverse formats like tutorials and blog posts for accessibility.
This resource is designed for NLP students and academic or industrial researchers who need a comprehensive, educational starting point to understand the field's core literature and evolution. It is particularly valuable for those new to NLP or looking to fill gaps in their knowledge of seminal and contemporary works.
Developers and researchers choose this over manually searching for papers because it offers a carefully selected, evolving list that balances historical foundations with recent breakthroughs, saving significant curation time. Its inclusion of non-peer-reviewed tutorials and surveys makes complex topics more accessible than raw paper collections.
100 Must-Read NLP Papers
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Compiled by Masato Hagiwara with input from Quora contributors, ensuring a vetted selection of influential papers that have stood the test of time or sparked major trends.
Includes tutorials, surveys, and blog posts alongside peer-reviewed papers, as stated in the README, making complex topics more accessible to learners without deep academic backgrounds.
Organized into categories like machine learning, neural models, and machine translation, providing a clear educational trajectory from foundations to advanced concepts.
Actively maintained through GitHub pull requests and issues, allowing the list to incorporate new important publications based on collective feedback.
The README explicitly states the list is 'far from complete or objective,' relying on curator bias and community input, which can omit key papers or over-represent certain areas.
Papers are listed without annotations, ratings, or difficulty levels, forcing users to independently assess relevance and content, which can be time-consuming for beginners.
Lacks interactive features like search, filtering, or personalized recommendations, making it cumbersome to navigate for specific research needs or quick reference.