Dataset quantization

DQ · ICCV 2023

Compresses large datasets into subsets usable by any architecture, reporting training with little or no loss from 60% of ImageNet and 20% of Alpaca instruction data across vision and language tasks.

Adaptive Dataset Quantization

Muquan Li, Dongyang Zhang, Qiang Dong et al. · AAAI 2025notablepaper ↗

Replaces Dataset Quantization's uniform bin sampling with an adaptive strategy that scores each generated bin's representativeness (texture level), diversity (contrastive-learning based) and importance, then samples accordingly; reports an average 3% accuracy gain over DQ across benchmarks with better cross-architecture generalization.

Mirage — Mirage: Model-Agnostic Graph Distillation for Graph Classification

Mridul Gupta, Sahil Manchanda, Hariprasad Kodamana et al. · ICLR 2024notableGraphspaper ↗code ↗

Mirage condenses a graph-classification dataset by decomposing every input graph into the multiset of computation trees a message-passing GNN would build on it, then compressing the (typically skewed) frequency distribution of these computation trees into a concise summary, rather than emulating gradient flows on the full dataset. Because it needs no full-dataset training and no target architecture during distillation, Mirage is unsupervised and architecture-agnostic, and the paper reports better generalization accuracy, compression and distillation efficiency than prior graph distillation baselines.

ProtoBlend — Efficient Video Dataset Distillation via Cluster-Guided Prototype Blending

Chongle Ren, Guang Li, Wenbo Huang et al. · arXiv 2026Videopaper ↗

Builds distilled video sets without any gradient-based optimization of the stored clips: a teacher model selects one high-confidence temporal segment per source video, cluster-guided allocation partitions the selected clips per class in the teacher's feature space and assigns one distilled slot per cluster, and each slot is blended with an in-cluster anchor clip while their teacher predictions are mixed with the same coefficient. Reports a competitive accuracy-efficiency trade-off against iteratively optimized video distillation methods on four trimmed action-recognition benchmarks.

C2TC — C$^{2}$TC: A Training-Free Framework for Efficient Tabular Data Condensation

Sijia Xu, Fan Li, Xiaoyang Wang et al. · arXiv 2026Other datapaper ↗

Condenses tabular data without any gradient-based training by reformulating condensation as class-adaptive cluster allocation: a heuristic local search alternates between soft cluster assignment and class-wise clustering, and a hybrid categorical encoding keeps the clustering meaningful over heterogeneous discrete features. Reports at least two orders of magnitude faster condensation than prior gradient-based tabular baselines across 10 real-world datasets, with better downstream accuracy.

Dataset Color Quantization: A Training-Oriented Framework for Dataset-Level Compression

Chenyue Yu, Lingao Xiao, Jinhong Deng et al. · ICLR 2026paper ↗

Reduces color-space redundancy across a dataset by enforcing consistent palettes across similar images, selectively retaining semantically important colors under model-perception guidance, and preserving structural detail needed for feature learning; reports improved training performance under aggressive compression on CIFAR-10/100, Tiny-ImageNet and ImageNet-1K.

IDTD — Video Set Distillation: Information Diversification and Temporal Densification

Yinjie Zhao, Heng Zhao, Bihan Wen et al. · arXiv 2024Videopaper ↗

This paper introduces video set distillation, jointly reducing within-sample redundancy (across frames of one video) and inter-sample redundancy (across videos in a set), which prior key-frame selection, dataset pruning or single-video distillation methods each address only one of. Its IDTD method uses a Feature Pool and Feature Selectors to preserve diversity across samples and a Temporal Fusor to keep temporal information dense within each synthesized video, reporting state-of-the-art results in video dataset distillation.

Dataset Quantization with Active Learning based Adaptive Sampling

Zhenghao Zhao, Yuzhang Shang, Junyi Wu et al. · ECCV 2024paper ↗code ↗

Shows that uniform per-class sample counts are unnecessary for dataset quantization -- some classes tolerate large reductions with little performance loss -- and proposes an active-learning-based adaptive sampling strategy (DQAS) that reallocates samples from stable to sensitive classes, plus a revised pipeline that uses final-stage quantization features to build more precise bins; outperforms prior dataset compression methods across multiple datasets.