flâneur — a map of the web's best reading

Generalized Language Models | Lil'Log

lilianweng.github.io · 7,830 words · saved by 1 readers

[Updated on 2019-02-14: add ULMFiT and GPT-2.] [Updated on 2020-02-29: add ALBERT.] [Updated on 2020-10-25: add RoBERTa.] [Updated on 2020-12-13: add T5.] [Updated on 2020-12-30: add GPT-3.] [Updated on 2021-11-13: add XLNet, BART and ELECTRA; Also updated the Summary section.] Fig. 0. I guess they are Elmo & Bert? (Image source: here) We have seen amazing progress in NLP in 2018. Large-scale pre-trained language modes like OpenAI GPT and BERT have achieved great performance on a variety of language tasks using generic model architectures.

Table of Contents CoVe NMT Recap Use CoVe in Downstream Tasks ELMo Bidirectional Language Model ELMo Representations Use ELMo in Downstream Tasks Cross-View Training Model Architecture Multi-Task Learning Use CVT in Downstream Tasks ULMFiT GPT Transformer Decoder as Language Model Byte Pair Encoding Supervised Fine-Tuning BERT Pre-training Tasks Input Embedding Use BERT in Downstream Tasks ALBERT Factorized Embedding Parameterization Cross-layer Parameter Sharing Sentence-Order Prediction (SOP) GPT-2 Zero-Shot Transfer BPE on Byte Sequences Model Modifications RoBERTa T5 GPT-3 XLNet BART ELECT

Explore this link on the map →

saved by

related reading