Toward Robust In-Context Learning: Leveraging Out-of-distribution Proxies for Target Inaccessible Demonstration Retrieval
Toward Robust In-Context Learning: Leveraging Out-of-distribution Proxies for Target Inaccessible Demonstration Retrieval
迈向鲁棒的上下文学习：利用分布外代理进行目标不可访问的示例检索
Abstract: Although studies have demonstrated that Large Language Models (LLMs) can perform well on Out-of-Distribution (OOD) tasks, their advantage tends to diminish as the distribution shift becomes more severe. Consequently, researchers aim to retrieve distributionally similar and informative demonstrations from the available source domain to boost the inference capabilities of LLMs. 摘要： 尽管研究表明大型语言模型（LLMs）在分布外（OOD）任务上表现良好，但随着分布偏移变得更加严重，其优势往往会减弱。因此，研究人员旨在从可用的源域中检索分布相似且信息丰富的示例，以增强 LLMs 的推理能力。
However, in practical scenarios where the target domain is inaccessible, evaluating the unknown distribution is challenging, which indirectly impacts the quality of the selected demonstrations. To address this problem, we propose DOPA, a demonstration search framework that incorporates an OOD proxy to approximate the inaccessible target domain and guide the retrieval process. 然而，在目标域不可访问的实际场景中，评估未知分布具有挑战性，这会间接影响所选示例的质量。为了解决这个问题，我们提出了 DOPA，这是一个示例搜索框架，它结合了 OOD 代理来近似不可访问的目标域并指导检索过程。
Building on proxy-based evaluation, DOPA further introduces a Mahalanobis distance-based global diversity constraint to ensure sufficient diversity among the retrieved demonstrations. Experimental results on multiple LLMs and tasks demonstrate that DOPA effectively enhances robustness in OOD settings. 在基于代理的评估基础上，DOPA 进一步引入了基于马氏距离（Mahalanobis distance）的全局多样性约束，以确保检索到的示例之间具有足够的多样性。在多个 LLMs 和任务上的实验结果表明，DOPA 有效地增强了 OOD 设置下的鲁棒性。
Paper Details:
- Authors: Hao Xu, Rite Bo, Fausto Giunchiglia, Yingji Li, Rui Song
- Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
- arXiv ID: 2606.00014
论文详情：
- 作者： Hao Xu, Rite Bo, Fausto Giunchiglia, Yingji Li, Rui Song
- 学科： 计算与语言 (cs.CL)；人工智能 (cs.AI)
- arXiv ID： 2606.00014