通过切割统计保障源自由域适应中的相互纠正
Safeguarding Mutual Correction in Source-Free Domain Adaptation via Cut Statistics
浏览论文内容
中文总结 AI 辅助
针对源自由域适应中模型相互纠正易受错误传播影响的问题，提出SafeCut方法，利用切割统计作为无标签可靠性度量来门控跨模型监督，动态控制监督方向与强度，理论保证净正纠正信号，并在多个基准上取得最先进性能。
中文摘要 AI 辅助
源自由域适应（SFDA）旨在无需访问原始源域的情况下，将源预训练模型适应到未标记的目标域。早期的单模型方法依赖于自我精炼，但本质上容易受到确认偏差的影响，难以纠正自身的系统性错误。为克服这一局限，近期方法引入视觉-语言（ViL）模型作为外部知识源。然而，这些方法主要采用单向范式，即主要利用ViL模型来监督源预训练模型。这忽略了一个关键的结构特性：两种模型表现出不同的失败模式——当其中一个产生错误预测时，另一个可能产生正确预测，从而在目标域内创造了相互纠正的自然机会。然而，在没有真实标签的情况下，识别哪个模型在给定样本上是正确的并非易事，而天真地交换预测可能会在模型间传播错误。为应对这一挑战，我们提出了SafeCut，一种新颖的方法，利用切割统计作为预测可靠性的无标签度量，来门控跨模型监督。我们的方法基于相对可靠性动态控制监督的方向和强度，在逐样本基础上选择性地放大真实纠正，同时抑制错误纠正。我们进一步提供了理论论证，表明这种可靠性门控机制保证了净正纠正信号。在多个多样的SFDA基准上的大量实验表明，SafeCut达到了最先进的性能，突显了通过切割统计保障SFDA中相互纠正的有效性。
英文摘要
Source-Free Domain Adaptation (SFDA) aims to adapt a source-pretrained model to an unlabeled target domain without access to the original source domain. While early single-model approaches rely on self-refinement, they are inherently susceptible to confirmation bias and struggle to correct their own systematic errors. To overcome this limitation, recent methods introduce Vision-Language (ViL) models as external knowledge sources. However, these approaches operate in a largely unidirectional paradigm, using the ViL model primarily to supervise the source-pretrained model. This overlooks a key structural property: the two models exhibit distinct failure modes -- where one produces an incorrect prediction, the other may produce a correct one, creating a natural opportunity for mutual correction within the target domain. Yet, without ground-truth labels, identifying which model is correct on any given sample is non-trivial, and naively exchanging predictions risks propagating errors across models. To address this challenge, we propose SafeCut, a novel approach that leverages the cut statistic as a label-free measure of prediction reliability to gate cross-model supervision. Our approach dynamically controls both the direction and strength of supervision based on relative reliability, selectively amplifying true corrections while suppressing miscorrections on a per-sample basis. We further provide theoretical justification showing that this reliability-gated mechanism guarantees a net-positive correction signal. Extensive experiments across diverse SFDA benchmarks demonstrate that SafeCut achieves state-of-the-art performance, highlighting the effectiveness of safeguarding mutual correction in SFDA via cut statistics.
发表机构
- Korea University（高丽大学）
机构由 AI 辅助整理，请以论文原文为准。
