Compară metode

	Mecanismul de atenție	Ajustarea fină a modelelor GPT	Atenție Multi-Capete (Multi-Head Self-Attention)
Domeniu	Învățare profundă	Învățare profundă	Învățare profundă
Familie	Machine learning	Machine learning	Machine learning
Anul apariției	2015	2019	2017
Autorul original	Bahdanau, D.; Luong, M.T.	Radford, A. et al. (OpenAI)	Vaswani, A. et al.
Tip	Neural attention layer (encoder-decoder)	Fine-tuning of pretrained autoregressive language models	Attention mechanism (Transformer core)
Sursa seminală	Bahdanau, D., Cho, K. & Bengio, Y. (2015). Neural Machine Translation by Jointly Learning to Align and Translate. ICLR.	Radford, A., Wu, J., Child, R., Luan, D., Amodei, D. & Sutskever, I. (2019). Language Models are Unsupervised Multitask Learners. OpenAI Technical Report.	Vaswani, A. et al. (2017). Attention Is All You Need. NeurIPS.
Denumiri alternative	Dikkat Mekanizması (Bahdanau / Luong Attention), dikkat mekanizmasi, neural attention, additive attention	GPT İnce Ayar ve Talimat Uyarlaması, GPT fine-tuning, instruction tuning, LLM fine-tuning	Öz-Dikkat ve Çok Başlı Dikkat (Multi-Head Self-Attention), öz-dikkat, multi-head attention, scaled dot-product attention
Înrudite	5	5	5
Rezumat	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.	GPT fine-tuning adapts pretrained autoregressive language models such as GPT-2/3/4 or LLaMA — introduced in OpenAI's 2019 work by Radford and colleagues — to domain-specific data or to instruction following via reinforcement learning from human feedback (RLHF) or DPO. It is used for instruction following, domain adaptation, and generative tasks.	Multi-head self-attention, introduced by Vaswani and colleagues in 2017, is the mechanism that lets every position in a sequence compute its relationship to all other positions in parallel. It is the core of the Transformer architecture and the foundation underneath BERT, GPT, and T5.