第43届国际机器学习大会（The 43rd International Conference on Machine Learning， ICML 2026）于2026年7月6日至11日在韩国首尔举行。ICML是机器学习领域历史最悠久、最具国际影响力的顶级学术会议之一，与NeurIPS、ICLR并称为机器学习“三大顶会”，被中国计算机学会（CCF）推荐为A类会议，会议主题覆盖机器学习理论、算法优化、深度学习、强化学习、生成模型及大模型应用等前沿方向，并邀请了多位国际知名学者进行主题演讲。

ICML 2026大会

实验室博士生郑威参加了ICML 2026大会，论文《Unsupervised Process-Aware Coreset Selection for In-Context Learning》被大会录用。参会期间与各国学者进行了学术交流，展示了研究成果，并聆听了多场报告，收获颇丰。

论文摘要如下：

We address the challenge of unsupervised coreset selection for few-shot in-context learning (ICL). The goal is to select a small subset of examples under a fixed annotation budget to yield effective prompts for large language models. Existing geometry-based methods often yield coresets that suffer from a skewed distribution, due to the oversampling of peripheral examples and high local redundancy. To address these issues, we propose a process-aware framework for coreset selection. It jointly optimizes the diversity and representativeness of selected samples via an adaptive submodular objective. It ensures representativeness by selecting samples based on local density awareness, while promoting diversity by imposing a redundancy penalty relative to the evolving selected set. Thus, it performs process-aware balancing of representativeness and diversity based on the selection context. Extensive experiments on 7 NLP datasets demonstrate that our method consistently outperforms state-of-the-art coreset selection methods in downstream ICL performance. Further analysis validates that our approach better balances diversity and representativeness throughout the selection process, while retaining the theoretical guarantees of adaptive submodular optimization.

论文海报