Structured Prompting: Scaling In-Context Learning to 1,000 Examples
This addresses the problem of scaling in-context learning for users of large language models, representing a novel method for a known bottleneck rather than incremental.
The paper tackles the limitation of conventional in-context learning in large language models, which is restricted by length constraints, by introducing structured prompting to scale it to thousands of examples, resulting in improved end-task performance and reduced evaluation variance as demonstration examples increase.
Large language models have exhibited intriguing in-context learning capability, achieving promising zero- and few-shot performance without updating the parameters. However, conventional in-context learning is usually restricted by length constraints, rendering it ineffective to absorb supervision from a large number of examples. In order to go beyond few shots, we introduce structured prompting that breaks the length limit and scales in-context learning to thousands of examples. Specifically, demonstration examples are separately encoded with well-designed position embeddings, and then they are jointly attended by the test example using a rescaled attention mechanism. So we can scale the number of exemplars with linear complexity instead of quadratic complexity with respect to length. Experimental results on a diverse set of tasks show that our approach improves end-task performance and reduces evaluation variance over conventional in-context learning as the number of demonstration examples increases. Code has been released at https://aka.ms/structured-prompting.