发表机构
Mohammed Premier University; Sultan Moulay Slimane University（穆罕默德一世大学; 苏丹穆莱·斯利曼大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对演化语义概念偏移问题，本文提出SemReWrite框架，结合低秩重写机制与结构化语义记忆等，构建EvoShift-Bench基准并引入多评价指标，在视觉持续学习中实现了修订语义与保留知识的更优平衡。
AI 中文摘要
视觉基础模型通常在假设输入数据的外观可能发生变化但预测任务的语义含义保持固定的前提下进行适配。然而，在长时视觉系统中，分类法、策略和概念定义本身可能会发生演化，导致相同的视觉证据需要不同的解释。我们将这种设置称为演化语义概念偏移，并引入SemReWrite，这是一个用于选择性更新过时视觉-语义映射同时保留有效知识的框架。SemReWrite表示旧语义规范与修订后语义规范之间的变化，将语义差异与稀疏修订后的监督结合以定位受影响的视觉区域，并使用依赖输入的低秩重写机制，结合结构化语义记忆、保留和过时决策抑制。我们进一步引入EvoShift-Bench，涵盖ImageNet、iNaturalist、CUB-200-2011和DomainNet，其语义转变包括类拆分、合并、边界修订、插入、部分重新定义、重现以及混合语义-外观偏移。为了明确评估选择性语义修订，我们分别引入了针对受影响区域的重写准确率（RA）和针对未受影响区域的保留准确率（PA）、用于测量残留过时语义关联的过时保留率（OR），以及联合总结重写和保留性能的选择性修订得分（SRS）。实验表明，与提示替换、传统微调、参数高效适配和持续学习策略相比，SemReWrite在学习修订后语义与保留未受影响知识之间实现了更优的平衡。
英文摘要
Visual foundation models are commonly adapted under the assumption that the appearance of incoming data may change while the semantic meaning of the prediction task remains fixed. In long-lived visual systems, however, taxonomies, policies, and concept definitions can themselves evolve, causing the same visual evidence to require a different interpretation. We study this setting as evolving semantic concept shift and introduce SemReWrite, a framework for selectively updating obsolete visual--semantic mappings while preserving knowledge that remains valid. SemReWrite represents changes between old and revised semantic specifications, combines semantic discrepancy with sparse revised supervision to localize affected visual regions, and uses an input-dependent low-rank rewriting mechanism together with structured semantic memory, preservation, and obsolete-decision suppression. We further introduce EvoShift-Bench, spanning ImageNet, iNaturalist, CUB-200-2011, and DomainNet, with semantic transitions including class split, merge, boundary revision, insertion, partial redefinition, recurrence, and mixed semantic--appearance shift. To explicitly evaluate selective semantic revision, we introduce Rewrite Accuracy (RA) and Preservation Accuracy (PA) for affected and unaffected regions, respectively, Obsolete Retention (OR) for measuring residual outdated semantic associations, and the Selective Revision Score (SRS), which jointly summarizes rewriting and preservation performance. Experiments show that SemReWrite achieves a stronger balance between learning revised semantics and retaining unaffected knowledge than prompt replacement, conventional fine-tuning, parameter-efficient adaptation, and continual-learning strategies.
