Glossary term
Glossary term
Agentic Systems
A model that infers a prediction based on its own previous predictions. For example, auto-regressive language models predict the next token based on the previously predicted tokens. All Transformer-based large language models are auto-regressive.
In contrast, GAN-based image models are usually not auto-regressive since they generate an image in a single forward-pass and not iteratively in steps. However, certain image generation models are auto-regressive because they generate an image in steps.
Created for this library
A code completion vendor uses an auto-regressive model that produces tokens left to right, allowing the IDE to stream suggestions as the user types.
A speech synthesis startup deploys an auto-regressive waveform model in its product but caches outputs for high-volume phrases to control latency.
A trading research team trains an auto-regressive model on hourly returns to forecast the next-step price for a basket of equity index futures.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License