用于主题驱动的个性化文本到图像生成的阶段感知适应与分布校准
Stage-Aware Adaptation and Distribution Calibration for Subject-Driven Personalized Text-to-Image Generation
AI总结：
研究主题驱动的个性化文本到图像生成问题，提出将其分解为训练端的 SPaRa 和推理端的 DCAL 两个互补组件的框架 SPaRa-DCAL，通过理论分析和实验表明该框架能提升相关指标，并指出应综合多方面评估个性化生成。
摘要：
主题驱动的个性化文本到图像生成需要预训练的扩散模型从一些参考图像中获取特定主题，同时保留主题身份、遵循新颖文本提示并保持样本多样性。现有基于优化的方法通过全量微调、文本嵌入优化或低秩参数更新来实现主题适应；PaRa 从参数秩降低的角度进一步约束个性化。然而，统一的低秩约束或统一的适配器强度无法明确区分不同去噪阶段的能力要求。此外，主要由身份相似性驱动的推理时候选选择可能会在视觉表示空间中压缩所选样本。我们将问题分解为两个互补组件：SPaRa 表示训练端的阶段感知低秩适应，DCAL 表示推理端的分布校准候选选择，SPaRa-DCAL 表示组合框架。理论分析表明，时间步依赖缩放控制低秩适配器的有效扰动幅度，而身份偏差候选选择在明确条件下限制所选特征围绕参考中心的半径。在 SDXL 和 DreamBooth 30 主题协议下的可审计实验表明，DCAL 在固定的 LoRA 候选池上提高了 1-LPIPS、CLIP-I、DINO-I 和 CLIP-T，同时揭示了与 CLIP/DINO 成对多样性和成对 LPIPS 的明显权衡。这些结果表明，个性化生成应通过身份一致性、文本对齐和表示多样性而非仅身份度量来评估。
英文摘要：
Subject-driven personalized text-to-image generation requires a pretrained diffusion model to acquire a specific subject from a few reference images while preserving subject identity, following novel text prompts, and maintaining sample diversity. Existing optimization-based methods instantiate subject adaptation through full fine-tuning, textual embedding optimization, or low-rank parameter updates; PaRa further constrains personalization from the perspective of parameter rank reduction. However, a uniform low-rank constraint or a uniform adapter strength cannot explicitly distinguish the capacity requirements of different denoising stages. Moreover, inference-time candidate selection driven mainly by identity similarity may compress the selected samples in the visual representation space. We decompose the problem into two complementary components: SPaRa denotes training-side stage-aware low-rank adaptation, DCAL denotes inference-side distribution-calibrated candidate selection, and SPaRa-DCAL denotes the combined framework. Theoretical analysis shows that timestep-dependent scaling controls the effective perturbation magnitude of a low-rank adapter, while identity-biased candidate selection restricts the radius of selected features around the reference center under explicit conditions. Auditable experiments under the SDXL and DreamBooth 30-subject protocol show that DCAL improves 1-LPIPS, CLIP-I, DINO-I, and CLIP-T on a fixed LoRA candidate pool, while revealing a clear trade-off with CLIP/DINO pairwise diversity and pairwise LPIPS. These results indicate that personalized generation should be evaluated through identity consistency, text alignment, and representation diversity rather than identity metrics alone.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
