English

Fine-tune Language Models to Approximate Unbiased In-context Learning

Machine Learning 2023-10-06 v1

Abstract

In-context learning (ICL) is an astonishing emergent ability of large language models (LLMs). By presenting a prompt that includes multiple input-output pairs as examples and introducing a new query input, models can generate the corresponding output. However, the performance of models heavily relies on the quality of the input prompt when implementing in-context learning. Biased or imbalanced input prompts can significantly degrade the performance of language models. To address this issue, we introduce a reweighted algorithm called RICL (Reweighted In-context Learning). This algorithm fine-tunes language models using an unbiased validation set to determine the optimal weight for each input-output example to approximate unbiased in-context learning. Furthermore, we also introduce a low-cost reweighted algorithm, a linear optimal weight approximation algorithm called LARICL (Linear Approximation of Reweighted In-context Learning). This algorithm requires minimal training cost while providing effective results. We prove the convergence of our algorithm and validate its performance through experiments conducted on a numerical dataset. The experimental findings reveal a substantial improvement in comparison to benchmarks including the performance of casual prompt-based in-context learning and the performance of a classic fine-tuning method.

Keywords

Cite

@article{arxiv.2310.03331,
  title  = {Fine-tune Language Models to Approximate Unbiased In-context Learning},
  author = {Timothy Chu and Zhao Song and Chiwun Yang},
  journal= {arXiv preprint arXiv:2310.03331},
  year   = {2023}
}
R2 v1 2026-06-28T12:41:10.289Z