2025/11/13 by Brian B. Moser, Moser, Brian B., Sarode, Shalini +13
Computer Science · #Advanced Neural Network Applications #Artificial Intelligence (cs.AI) #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning and Data Classification
paper · pdf · doi:10.48550/arxiv.2511.09905
openalex publication_date 2025/11/13 · openalex created_date 2025/11/15 · openalex updated_date 2026/07/28
Dataset distillation (DD) promises compact yet faithful synthetic data, but existing approaches often inherit the inductive bias of a single teacher model. As dataset size increases, this bias drives generation toward overly smooth, homogeneous samples, reducing intra-class diversity and limiting generalization. We present PRISM (PRIors from diverse Source Models), a framework that disentangles architectural priors during synthesis. PRISM decouples the logit-matching and regularization objectives, supervising them with different teacher architectures: a primary model for logits and a stochastic subset for batch-normalization (BN) alignment. On ImageNet-1K, PRISM consistently and reproducibly outperforms single-teacher methods (e.g., SRe2L) and recent multi-teacher variants (e.g., G-VBSM) at low- and mid-IPC regimes. The generated data also show significantly richer intra-class diversity, as reflected by a notable drop in cosine similarity between features. We further analyze teacher selection strategies (pre- vs. intra-distillation) and introduce a scalable cross-class batch formation scheme for fast parallel synthesis. Code will be released after the review period.