Nlp Glossary
Browse 64 nlp terms defined in plain English, from the cultural dictionary of computing.
64 Nlp Terms
- Attention
- The core mechanism of transformer models that allows each token to dynamically attend to (weigh the importance of) every other token in the sequence....
- Beam Search
- A decoding strategy that keeps the top-scoring partial sequences instead of only the single best candidate. It influences how models are trained, evaluated, or...
- BERT
- A bidirectional transformer language model pre-trained with masked language modeling and next sentence objectives. It influences how models are trained,...
- Byte Pair Encoding
- A tokenization algorithm that repeatedly merges common symbol pairs to build a useful subword vocabulary. It influences how models are trained, evaluated, or...
- Causal Language Model
- A language model trained to predict the next token using only preceding context. It influences how models are trained, evaluated, or served, and it can...
- Chatbot
- A program that simulates conversation with human users. Ranges from ELIZA's 1966 pattern-matching parlor tricks to modern LLM-powered assistants that can debug...
- ELMo
- Embeddings from Language Models, a contextual word representation model built from bidirectional language models. It influences how models are trained,...
- Embedding Model
- A neural network trained to convert text, images, or other data into fixed-dimensional vectors (embeddings) that capture semantic meaning. Similar items...
- GLUE Benchmark
- A benchmark suite of language understanding tasks used to compare NLP model performance. It influences how models are trained, evaluated, or served, and it can...
- Large Language Model
- A neural network with billions of parameters trained on massive text datasets to understand and generate human language. GPT, Claude, Llama, and Gemini are...
- LLM
- Large Language Model — a neural network trained on vast amounts of text data that can generate, summarize, translate, and reason about human language.
- LLM Agent
- LLM Agent is an AI or ML concept used to represent, train, evaluate, or deploy learned systems. It is commonly used for building production models and research...
- LLM Evaluation
- LLM Evaluation is an evaluation concept used to measure model quality, robustness, or efficiency. It is commonly used for comparing systems before release or...
- LLM Fine-Tuning
- LLM Fine-Tuning is a stage of model optimization where weights or behaviors are adjusted from data or feedback. It is commonly used for foundation-model...
- LLM Inference
- LLM Inference is the phase where a trained model processes new inputs to produce predictions or generations. It is commonly used for production APIs, batch...
- LLM Routing
- LLM Routing is an AI or ML concept used to represent, train, evaluate, or deploy learned systems. It is commonly used for building production models and...
- LLM Safety
- LLM Safety is a safety or governance control for limiting harmful, noncompliant, or insecure model behavior. It is commonly used for production AI systems that...
- Local LLM
- Local LLM is an AI or ML concept used to represent, train, evaluate, or deploy learned systems. It is commonly used for building production models and research...
- Long Context
- Long Context is an AI or ML concept used to represent, train, evaluate, or deploy learned systems. It is commonly used for building production models and...
- Masked Language Model
- Masked Language Model is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building neural...
- Multimodal Embedding
- Multimodal Embedding is a dense numerical representation that places semantically related items near each other in vector space. It is commonly used for...
- Multimodal Tokenizer
- Multimodal Tokenizer is a unit or control marker used when text is segmented and generated by language models. It is commonly used for prompt assembly,...
- Named Entity Recognition
- Named Entity Recognition is an application task where a model extracts structured meaning or predictions from raw input. It is commonly used for NLP and vision...
- Natural Language Generation
- Natural Language Generation is a generative modeling concept for producing new content such as text, images, audio, or video. It is commonly used for creative...
- Natural Language Inference
- Natural Language Inference is the phase where a trained model processes new inputs to produce predictions or generations. It is commonly used for production...
- Natural Language Processing
- Natural Language Processing is an evaluation concept used to measure model quality, robustness, or efficiency. It is commonly used for comparing systems before...
- Natural Language Understanding
- Natural Language Understanding is an application task where a model extracts structured meaning or predictions from raw input. It is commonly used for NLP and...
- Neural Machine Translation
- Neural Machine Translation is an application task where a model extracts structured meaning or predictions from raw input. It is commonly used for NLP and...
- N-Gram
- N-Gram is an AI or ML concept used to represent, train, evaluate, or deploy learned systems. It is commonly used for building production models and research...
- OpenAI
- OpenAI is an AI company and research lab that develops foundation models, developer APIs, and products such as ChatGPT. It is commonly used for building...
- Output Token
- Output Token is a unit or control marker used when text is segmented and generated by language models. It is commonly used for prompt assembly, decoding...
- Perplexity
- A metric that measures how well a probability model predicts a sample. Lower perplexity means the model is less 'surprised' by the data. Informally, it...
- Positional Embedding
- Positional Embedding is a dense numerical representation that places semantically related items near each other in vector space. It is commonly used for...
- Prompt Caching
- Prompt Caching is a prompting technique or prompt-side control used to shape large-model behavior at inference time. It is commonly used for chat systems,...
- Prompt Chaining
- Prompt Chaining is a prompting technique or prompt-side control used to shape large-model behavior at inference time. It is commonly used for chat systems,...
- Prompt Engineering Detail
- Prompt Engineering Detail is a prompting technique or prompt-side control used to shape large-model behavior at inference time. It is commonly used for chat...
- Prompt Injection Detail
- Prompt Injection Detail is a prompting technique or prompt-side control used to shape large-model behavior at inference time. It is commonly used for chat...
- Prompt Template
- Prompt Template is a prompting technique or prompt-side control used to shape large-model behavior at inference time. It is commonly used for chat systems,...
- Prompt Tuning
- Prompt Tuning is a prompting technique or prompt-side control used to shape large-model behavior at inference time. It is commonly used for chat systems,...
- Rope Embedding
- Rope Embedding is a dense numerical representation that places semantically related items near each other in vector space. It is commonly used for retrieval,...
- Rotary Position Embedding
- Rotary Position Embedding is a dense numerical representation that places semantically related items near each other in vector space. It is commonly used for...
- Semantic Search
- Search based on the meaning of queries rather than keyword matching. Converts both queries and documents into vector embeddings and finds the nearest matches...
- Sentence Embedding
- Sentence Embedding is a dense numerical representation that places semantically related items near each other in vector space. It is commonly used for...
- Sentence Transformer
- Sentence Transformer is an AI or ML concept used to represent, train, evaluate, or deploy learned systems. It is commonly used for building production models...
- Sentiment Analysis
- Sentiment Analysis is an application task where a model extracts structured meaning or predictions from raw input. It is commonly used for NLP and vision...
- Seq2Seq
- Seq2Seq is an AI or ML concept used to represent, train, evaluate, or deploy learned systems. It is commonly used for building production models and research...
- Sequence Modeling
- Sequence Modeling is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building neural...
- Sequence-to-Sequence
- Sequence-to-Sequence is an AI or ML concept used to represent, train, evaluate, or deploy learned systems. It is commonly used for building production models...
- Slot Filling
- A task where a system extracts specific fields or arguments from text, such as dates, names, or booking details.
- Stop Token
- Stop Token is a unit or control marker used when text is segmented and generated by language models. It is commonly used for prompt assembly, decoding control,...
- Subword Tokenization
- Subword Tokenization is a unit or control marker used when text is segmented and generated by language models. It is commonly used for prompt assembly,...
- System Prompt Detail
- System Prompt Detail is a prompting technique or prompt-side control used to shape large-model behavior at inference time. It is commonly used for chat...
- Text Embedding
- Text Embedding is a dense numerical representation that places semantically related items near each other in vector space. It is commonly used for retrieval,...
- Token
- The basic unit of text that a language model processes — roughly 3/4 of an English word on average. Text is split into tokens by a tokenizer before being fed...
- Token
- The basic unit of text processed by a language model — roughly ¾ of a word on average. 'Unbelievable' might be two tokens: 'un' and 'believable.' Token limits...
- Token Budget
- Token Budget is a unit or control marker used when text is segmented and generated by language models. It is commonly used for prompt assembly, decoding...
- Tokenization Detail
- Tokenization Detail is a unit or control marker used when text is segmented and generated by language models. It is commonly used for prompt assembly, decoding...
- Tokenizer
- The component that splits text into tokens before feeding it to a language model. Different tokenizers produce different tokens — 'unhappiness' might become...
- Token Limit
- Token Limit is a unit or control marker used when text is segmented and generated by language models. It is commonly used for prompt assembly, decoding...
- Tokens Per Second Detail
- Tokens Per Second Detail is a unit or control marker used when text is segmented and generated by language models. It is commonly used for prompt assembly,...
- Vector Embedding
- Vector Embedding is a dense numerical representation that places semantically related items near each other in vector space. It is commonly used for retrieval,...
- Word2Vec
- Word2Vec is an AI or ML concept used to represent, train, evaluate, or deploy learned systems. It is commonly used for building production models and research...
- Word Embedding
- Word Embedding is a dense numerical representation that places semantically related items near each other in vector space. It is commonly used for retrieval,...
- Zero-Shot Prompting
- Zero-Shot Prompting is a prompting technique or prompt-side control used to shape large-model behavior at inference time. It is commonly used for chat systems,...
Related Topics
- Llm (27 terms in common)
- Inference (25 terms in common)
- Ml (23 terms in common)
- Representation Learning (22 terms in common)
- Transformer (2 terms in common)
- Tokenization (2 terms in common)
- Evaluation (2 terms in common)
- Search (2 terms in common)
- Embeddings (2 terms in common)
- Transformers (1 terms in common)
- Architecture (1 terms in common)
- Bpe (1 terms in common)
- Deep Learning (1 terms in common)
- Preprocessing (1 terms in common)
- Vector (1 terms in common)
- Decoding (1 terms in common)
- Ai (1 terms in common)
- Foundation Model (1 terms in common)
- Vocabulary (1 terms in common)
- Information Extraction (1 terms in common)