ReBaPL：排斥式贝叶斯提示学习
ReBaPL: Repulsive Bayesian Prompt Learning
把提示词优化从点估计改成对多模态后验的采样，并用表征空间上的排斥力（MMD/Wasserstein）防止所有样本塌到同一模式；做视觉-语言下游适配时可直接套在现有提示学习框架上提升分布外泛化。
原文依据 a repulsive force derived from a potential function over probability metrics (including Maximum Mean Discrepancy and Wasserstein distance)
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论提示学习已成为微调大规模基础模型以适配下游任务的有效技术。然而，传统提示学习方法容易过拟合，且在分布外泛化上表现不佳。为解决这些局限，研究者提出了贝叶斯提示学习，将提示优化建模为贝叶斯推理问题以增强鲁棒性。本文提出排斥式贝叶斯提示学习（ReBaPL），一种新的贝叶斯提示学习方法，旨在高效探索提示复杂且往往多模态的后验景观。该方法将循环步长调度与随机梯度哈密顿蒙特卡洛（SGHMC）算法相结合，使探索阶段与利用阶段交替进行：前者用于发现新的模式，后者用于细化已有模式。此外，我们引入一种排斥力，它源自基于概率度量（包括最大均值差异与 Wasserstein 距离）的势函数，这些度量在不同提示所产生的表征分布上计算。这种表征空间中的排斥使探索多样化，并防止过早塌缩到单一模式。我们的方法能够更全面地刻画提示后验分布，从而提升泛化能力。与先前的贝叶斯提示学习方法不同，我们的方法为任何基于最大似然估计的现有提示学习方法提供了模块化的即插即用贝叶斯扩展。我们在多个基准数据集上验证了 ReBaPL 的有效性，其性能优于最优的提示学习方法。
英文原文摘要 1685 字符
Prompt learning has emerged as an effective technique for fine-tuning large-scale foundation models for downstream tasks. However, conventional prompt learning methods are prone to overfitting and can struggle with out-of-distribution generalization. To address these limitations, Bayesian prompt learning has been proposed, which frames prompt optimization as a Bayesian inference problem to enhance robustness. This paper introduces Repulsive Bayesian Prompt Learning (ReBaPL), a novel method for Bayesian prompt learning, designed to efficiently explore the complex and often multimodal posterior landscape of prompts. Our method integrates a cyclical step-size schedule with a stochastic gradient Hamiltonian Monte Carlo (SGHMC) algorithm, enabling alternating phases of exploration to discover new modes, and exploitation to refine existing modes. Furthermore, we introduce a repulsive force derived from a potential function over probability metrics (including Maximum Mean Discrepancy and Wasserstein distance) computed on the distributions of representations produced by different prompts. This representation-space repulsion diversifies exploration and prevents premature collapse to a single mode. Our approach allows for a more comprehensive characterization of the prompt posterior distribution, leading to improved generalization. In contrast to prior Bayesian prompt learning methods, our method provides a modular plug-and-play Bayesian extension of any existing prompt learning method based on maximum likelihood estimation. We demonstrate the efficacy of ReBaPL on several benchmark datasets, showing superior performance over state-of-the-art prompt learning methods.