Large language model (LLM) NLP
noun phrase
Definition:A class of large-scale, pre-trained neural language models, now predominantly transformer-based, trained on massive text corpora to perform language understanding and generation across a wide range of tasks. Recent survey literature defines LLMs as large-scale pre-trained statistical language models based on neural networks, and notes that they mainly refer to transformer-based models with very large parameter counts [IBM Think; Minae et al. 2025].
Example in context: “LLMs are large-scale, pre-trained, statistical language models based on neural networks.” [Minaee et al. 2024]
Synonym: LLM
Related terms: foundation model (broader), transformer language model, generative language model, pretrained language model