models

Training a smaller "student" model to mimic the behavior of a larger "teacher" model, transferring the teacher's knowledge into a more compact form. The student learns from the teacher's outputs (soft...

Also: Knowledge Distillation, Model Distillation, Teacher-Student Learning

The process of further training a pre-trained model on a specific dataset to adapt it for a particular task or domain. Fine-tuning updates model weights (unlike prompting, which only changes inputs), ...

Also: Fine-tuning, Model Fine-tuning, Transfer Learning

A large model trained on broad data at scale, designed to be adapted to a wide range of downstream tasks. Foundation models are "pre-trained" on general data (text, images, code) and then specialized ...

Also: Base Model, Pre-trained Model

Running a trained model on input data to produce output (predictions, generated text, classifications). Inference is the "using" phase of machine learning, as opposed to training. For LLMs, inference ...

Also: Model Inference, Prediction, Forward Pass

Fine-tuning a language model on datasets of instructions and responses to improve its ability to follow natural language instructions. Instruction- tuned models understand and execute commands like "S...

Also: Instruction Fine-tuning, Instruct Models

A neural network trained on massive text corpora to predict and generate natural language. "Large" refers to both the model size (billions of parameters) and training data (trillions of tokens). LLMs ...

Also: LLM, Language Model, Foundation Model

A model that can process and/or generate multiple types of data (modalities) such as text, images, audio, and video. Multimodal models understand relationships across modalities—they can describe imag...

Also: Vision-Language Model, VLM, Multimodal LLM

Reducing the numerical precision of model weights and/or activations (e.g., from 32-bit floats to 8-bit or 4-bit integers) to decrease memory usage and increase inference speed, with minimal quality l...

Also: Model Quantization, Weight Quantization, INT8/INT4

A training technique that aligns LLMs with human preferences by using human feedback to train a reward model, then optimizing the LLM against that reward. RLHF is how raw pre-trained models become hel...

Also: Reinforcement Learning from Human Feedback, Human Feedback Training, Preference Learning

The basic unit of text that an LLM processes, typically a subword piece rather than a full word or character. Tokens are what models actually see—text is converted to tokens for input and tokens are c...

Also: Subword, Token ID

A component that converts raw text into tokens (numerical IDs) that an LLM can process, and converts tokens back to text. Tokenizers define the vocabulary of a model and how text is segmented. The tok...

Also: Tokenization, Token Encoder

A neural network architecture based on self-attention mechanisms, introduced in "Attention Is All You Need" (2017). Transformers process sequences by allowing each position to attend to all other posi...

Also: Transformer Architecture, Attention Architecture