What is BPE (Byte Pair Encoding)?

The process of splitting text into tokens for model input.

Technology that converts text in images or scans into machine-readable text.

A decoding strategy that samples only from the k most likely next tokens.

The core training algorithm that updates model weights by propagating prediction errors backward through the network.

A model that encodes queries and documents into separate vectors so they can be compared quickly at scale.

A decoding algorithm that keeps the top several candidate sequences at each step to find higher-probability outputs.