Self-ICL: Zero-Shot In-Context Learning with Self-Generated Demonstrations

TL;DR

Self-ICL enables large language models to perform zero-shot in-context learning by generating pseudo-demonstrations internally, improving task performance without access to external demonstration data.

Contribution

Introduces Self-ICL, a novel framework that leverages LMs' intrinsic abilities to generate pseudo-demonstrations for zero-shot in-context learning, aligning with real-world usage.

Findings

Outperforms zero-shot baselines on BIG-Bench Hard tasks

Achieves comparable results to real demonstrations with chain-of-thought prompting

Validated effectiveness through extensive analyses

Abstract

Large language models (LLMs) have exhibited striking in-context learning (ICL) ability to adapt to target tasks with a few input-output demonstrations. For better ICL, different methods are proposed to select representative demonstrations from existing training corpora. However, such settings are not aligned with real-world practices, as end-users usually query LMs without access to demonstration pools. In this work, we introduce Self-ICL -- a simple framework which bootstraps LMs' intrinsic capabilities to perform zero-shot ICL. Given a test input, Self-ICL first prompts the model to generate pseudo-inputs. Next, the model predicts pseudo-labels for the pseudo-inputs via zero-shot prompting. Finally, we perform ICL for the test input with the pseudo-input-label pairs as demonstrations. Evaluation on 23 BIG-Bench Hard tasks shows Self-ICL outperforms zero-shot baselines on both average accuracy and head-to-head comparison. Moreover, with zero-shot chain-of-thought, Self-ICL achieves results comparable to using real demonstrations. Additionally, we conduct a range of analyses to validate Self-ICL's effectiveness and provide insights for its behaviors under different settings.