OD3 presents an optimization-free dataset distillation framework for object detection that reports new state-of-the-art accuracy on COCO and VOC at compression ratios from 0.25% to 5%.
Super-Samples from Kernel Herding
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
abstract
We extend the herding algorithm to continuous spaces by using the kernel trick. The resulting "kernel herding" algorithm is an infinite memory deterministic process that learns to approximate a PDF with a collection of samples. We show that kernel herding decreases the error of expectations of functions in the Hilbert space at a rate O(1/T) which is much faster than the usual O(1/pT) for iid random samples. We illustrate kernel herding by approximating Bayesian predictive distributions.
citation-role summary
baseline 1
extension 1
citation-polarity summary
verdicts
UNVERDICTED 8representative citing papers
The Sinkhorn treatment effect is a new entropic optimal transport measure of divergence between counterfactual distributions that admits first- and second-order pathwise differentiability, debiased estimators, and asymptotically valid tests for distributional treatment effects.
ContextualJailbreak uses evolutionary search over simulated primed dialogues with novel mutations to reach 90-100% attack success on open LLMs and transfers to some closed frontier models at 15-90% rates.
A graph-based unified dataset pruning framework that formulates pruning as MWCP, derives a greedy algorithm with formal approximation guarantees, and demonstrates substantial training acceleration on ImageNet.
CRUMB speeds up PFN inference on large tabular datasets by clustering queries and selecting MMD-matched context subsets, outperforming prior selection methods on the 51-dataset TabArena benchmark across three architectures while handling covariate drift.
D3S2 combines class-balanced mask selection with diffusion-guided image synthesis and two consistency losses to distill 1% datasets that yield 24.99% mIoU on ADE20K and 35.49% on COCO-Stuff, beating random selection.
LFM models exhibit stability to data reduction and capacity shrinkage that is tied to the flow matching objective, enabling reduced-data training and coarse-to-fine inference with over 2x speedup.
Generalizing two DPP-based Monte Carlo estimators to continuous domains provides variance rates of O(N^{-(1+1/d)}) for a fixed DPP method and O(1/N) for a tailored DPP method, along with new sampling algorithms.
citing papers explorer
-
OD3: Optimization-free Dataset Distillation for Object Detection
OD3 presents an optimization-free dataset distillation framework for object detection that reports new state-of-the-art accuracy on COCO and VOC at compression ratios from 0.25% to 5%.
-
D3S2: Diffusion-Guided Dataset Distillation for Semantic Segmentation
D3S2 combines class-balanced mask selection with diffusion-guided image synthesis and two consistency losses to distill 1% datasets that yield 24.99% mIoU on ADE20K and 35.49% on COCO-Stuff, beating random selection.