in-context learning
A model picking up a task from instructions and examples in its prompt, with no change to its weights.
In more detail
Everything learned this way lasts only for the call: the next request starts from nothing unless the examples are sent again. It is why few-shot prompting and retrieval work at all.
Related terms
- few-shot prompting — Including a few worked input and output examples in the prompt to show the model the task and format.
- fine-tuning — Further training an existing model on your own examples to change its behaviour, style or format.
Taught in
- Zero-Shot And Few-Shot — AI Engineering: LLM Apps, RAG & Agents
References
- Brown et al. (2020): Language Models are Few-Shot Learners
- Min et al. (2022): Rethinking the Role of Demonstrations