Сравнение на методи

	Генериране на естествен език	Генериране с разширение за извличане (Retrieval-Augmented Generation, RAG)
Област	Извличане на текст	Извличане на текст
Семейство	Process / pipeline	Process / pipeline
Година на възникване	1970s (rule-based origins); 2000s (probabilistic); 2017+ (neural/transformer era)	2020
Създател	Reiter & Dale (classical pipeline, 2000); Gatt & Krahmer (modern survey, 2018)	Lewis, Patrick et al. (Meta AI / Facebook AI Research)
Тип	NLP generative task — structured data to natural language	Hybrid retrieval + generation pipeline
Основополагащ източник	Gatt, A. & Krahmer, E. (2018). Survey of the State of the Art in Natural Language Generation: Core Tasks, Applications and Evaluation. Journal of Artificial Intelligence Research, 61, 65-170.	Lewis, P. et al. (2020). Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks. Advances in Neural Information Processing Systems (NeurIPS), 33, 9459-9474.
Други названия	NLG, data-to-text, text generation, Doğal Dil Üretimi (NLG)	RAG, retrieval-augmented LLM, grounded generation, Erişim Destekli Metin Üretimi (RAG)
Свързани	7	7
Резюме	Natural Language Generation (NLG) is the branch of natural language processing that automatically produces fluent, human-readable text from structured data, knowledge graphs, or semantic representations. Formalised in the classical pipeline by Reiter and Dale (2000) and surveyed comprehensively by Gatt and Krahmer (2018), NLG powers applications ranging from automated financial reporting and weather bulletins to data storytelling and conversational agents.	Retrieval-Augmented Generation (RAG) is a natural-language-processing pipeline introduced by Lewis et al. in 2020 that strengthens a large language model (LLM) with evidence fetched at inference time from an external knowledge base. Instead of relying solely on what a model memorised during training, RAG first retrieves the most relevant passages from a document index and then hands those passages to the LLM as context, grounding the generated answer in verifiable, up-to-date information. The approach reduces hallucination and allows domain-specific or time-sensitive knowledge to be injected without retraining the model.