Neural NLP

Neural approaches to natural language processing use deep learning architectures to solve tasks like machine translation, sequence labeling, question answering, and information extraction. Modern neural NLP is dominated by Transformer-based models like BERT, GPT, and T5.

Foundational architectures

• Attention Is All You Need — introduces the Transformer, the foundational architecture for modern neural NLP.

Related topics

• Transformers — the dominant architecture in neural NLP
• Sequence Modeling — broader context of learning from sequential text
• Attention mechanisms in NLP — core mechanism enabling modern neural NLP