ORBiS Tree

TOPIC · ENGLISH EDITION

Concepts

Core technical concepts used across modern AI systems.

EN

AI Agent

An AI system that selects and executes actions over multiple steps to pursue a goal.

EN

Attention

A mechanism that lets a neural network weigh the relevance of different positions when computing a representation.

EN

Context Window

The amount of tokenized input and generated text a model can consider within one inference context.

EN

Diffusion Model

A generative model that learns to reverse a gradual noise process to create data such as images.

EN

Embedding

A numerical vector representation used to encode semantic or structural properties of data.

EN

Fine-tuning

Additional training that adapts a pre-trained model to a narrower task, domain, or behavior.

EN

GPU

A highly parallel processor widely used to train and run modern machine-learning models.

EN

Inference

The process of running a trained model on new input to produce predictions, embeddings, or generated output.

EN

KV Cache

A cache of attention keys and values used to avoid recomputing earlier context during autoregressive generation.

EN

Knowledge Distillation

A training method in which a smaller student model learns from the outputs or representations of a larger teacher model.

EN

Latency and Throughput

Two core serving metrics describing response delay and the amount of work a system completes over time.

EN

LoRA

Low-Rank Adaptation, a parameter-efficient technique for adapting large neural networks.

EN

Mixture of Experts (MoE)

A model architecture that routes each input to a subset of specialized parameter blocks called experts.

EN

Multimodal AI

AI systems that process or generate more than one type of data, such as text, images, audio, or video.

EN

Quantization

A technique that represents model weights or activations with lower numerical precision to reduce resource use.

EN

RLHF

Reinforcement Learning from Human Feedback, a method for aligning model behavior with human preferences.

EN

Retrieval-Augmented Generation (RAG)

A method that retrieves external information and supplies it to a generative model as context.

EN

Tokenization

The process of converting text or other input into discrete units that a model can process.

EN

Transformer

A neural network architecture that processes sequences primarily through attention mechanisms.

EN

Vector Search

Search that retrieves items by similarity between numerical vector representations.