用自生成标注实现上千次上下文学习，性能超越真实标签
Scaling Laws for Many-Shot In-Context Learning with Self-Generated Annotations
- 用自动生成标注替代真实标签，构建可扩展的上下文学习框架
- 超过1000个示范时性能达最优，验证了规模化潜力
- 提出迭代伪标签方法，分类任务提升最高达6.8%
为降低高质量标注数据的成本，研究者提出用自生成标注替代真实标签进行上下文学习（ICL）。尽管已有方法在少样本场景表现良好，但难以扩展至多样本场景。本文基于半监督学习框架，构建包含标注生成、示范选择和上下文推理的流程，提出一种简单基线，在零样本、少样本和多样本设置下均优于使用真实标签的ICL。值得注意的是，该基线存在显著缩放规律：当示范数量超过1000时达到最佳性能。为进一步挖掘半监督ICL的多样本潜力，提出IterPSD方法，融合迭代优化与课程伪标签技术，在分类任务上实现最高6.8%的性能提升。
原文摘要 · Abstract (English)
The high cost of obtaining high-quality annotated data for in-context learning (ICL) has motivated the development of methods that use self-generated annotations in place of ground-truth labels. While these approaches have shown promising results in few-shot settings, they generally do not scale to many-shot scenarios. In this work, we study ICL with self-generated examples using a framework analogous to traditional semi-supervised learning, consisting of annotation generation, demonstration selection, and in-context inference. Within this framework, we propose a simple baseline that outperforms ground-truth ICL in zero-shot, few-shot, and many-shot settings. Notably, we observe a scaling law with this baseline, where optimal performance is achieved with more than 1,000 demonstrations. To fully exploit the many-shot capabilities of semi-supervised ICL, we introduce IterPSD, an iterative annotation approach that integrates iterative refinement and curriculum pseudo-labeling techniques from semi-supervised learning, yielding up to 6.8% additional gains on classification tasks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。