This repository is a curated reading list for machine translation (MT), primarily focusing on Neural Machine Translation (NMT). It serves researchers, engineers, and students interested in the field, providing a structured overview of key papers, tutorials, and surveys.

How It Works

The list categorizes papers across various MT sub-topics, from foundational Statistical Machine Translation (SMT) concepts like word-based and phrase-based models to advanced NMT architectures, attention mechanisms, training objectives, and specialized translation tasks like low-resource or speech translation. It prioritizes recent and influential NMT research.

Quick Start & Requirements

This is a reading list, not a software package. No installation or specific requirements are needed beyond a web browser to access the content.

Highlighted Details

• Comprehensive categorization of MT topics, including SMT, NMT, evaluation metrics, and specific challenges.
• Extensive "10 Must Reads" section highlighting seminal papers with citation counts.
• Detailed breakdown of NMT model architectures, attention mechanisms, and training techniques.
• Includes WMT competition winners and summaries of their system techniques.

Maintenance & Community

Maintained by the Tsinghua Natural Language Processing Group. Suggestions for additions and improvements are welcomed.

Licensing & Compatibility

The content itself is a curated list of academic papers and does not have a specific software license. Individual papers are subject to their respective publication licenses.

Limitations & Caveats

The list acknowledges it is incomplete and categorization may be subject to refinement. It focuses heavily on NMT, with less emphasis on older SMT techniques, which are better covered by other resources.