Concept

Role of the Adapter in BERT-based Encoder-Decoder Models

In a BERT-based encoder-decoder architecture, an adapter is an optional layer that serves as a bridge between the encoder and the decoder. Its primary function is to map the output representations generated by the BERT encoder into a format that is more suitable for the decoder to process. This helps align the output space of the pre-trained encoder with the input requirements of the decoder.

0

1

Updated 2026-04-18

Contributors are:

Who are from:

Tags

Ch.2 Generative Models - Foundations of Large Language Models

Foundations of Large Language Models

Computing Sciences

Foundations of Large Language Models Course

Ch.1 Pre-training - Foundations of Large Language Models