Linganisha mbinu

	Attention Mechanism	BERT Embeddings	Transformer wa Maono
Nyanja	Ujifunzaji wa Kina	Uchimbaji wa Matini	Ujifunzaji wa Kina
Familia	Machine learning	Process / pipeline	Machine learning
Mwaka wa asili	2015	2019	2021
Mwanzilishi	Bahdanau, D.; Luong, M.T.	Devlin, Chang, Lee & Toutanova (Google AI)	Dosovitskiy, A. et al.
Aina	Neural attention layer (encoder-decoder)	Contextual transformer text-representation method	Transformer architecture for images (self-attention over patches)
Chanzo asilia	Bahdanau, D., Cho, K. & Bengio, Y. (2015). Neural Machine Translation by Jointly Learning to Align and Translate. ICLR.	Devlin, J., Chang, M.-W., Lee, K. & Toutanova, K. (2019). BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. NAACL-HLT, 4171-4186.	Dosovitskiy, A. et al. (2021). An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale. ICLR.
Majina mbadala	Dikkat Mekanizması (Bahdanau / Luong Attention), dikkat mekanizmasi, neural attention, additive attention	contextual embeddings, transformer embeddings, BERT Tabanlı Metin Gömülmeleri	Görsel Transformer (ViT), görsel transformer, ViT, patch transformer for images
Zinazohusiana	5	4	5
Muhtasari	The attention mechanism, introduced by Bahdanau, Cho and Bengio in 2015 and refined by Luong, Pham and Manning the same year, lets a sequence decoder dynamically learn which of the encoder's outputs to focus on at each step. Before the Transformer, it substantially improved machine-translation quality by freeing models from compressing an entire input into a single fixed vector.	BERT-based text embeddings, introduced by Devlin and colleagues at Google AI in 2019, turn text into context-sensitive dense vectors using a bidirectional Transformer encoder. Because the meaning of a word shifts with its context, BERT produces richer representations than static methods such as Word2Vec or topic models like LDA.	The Vision Transformer (ViT), introduced by Dosovitskiy and colleagues in 2021, splits an image into fixed-size patches, treats those patches as a sequence, and applies the Transformer self-attention mechanism to image classification. Given enough training data, it surpasses convolutional neural networks (CNNs).