Decoding from Neural Models | EPFL Graph Search

Explores pretraining models like BERT, T5, and GPT, discussing their training objectives and applications in natural language processing.

Covers advanced transformer concepts, focusing on pretraining and decoding techniques in NLP.

Covers contextual representations in NLP, focusing on ELMO and BERT architectures and their applications in various tasks.

Covers sequence to sequence models, their architecture, applications, and the role of attention mechanisms in improving performance.

Explores neural model analysis in NLP, covering evaluation, probing, and ablation studies to understand model behavior and interpretability.