The paper investigates how the usefulness of training samples, as determined by coreset selection, depends on the learner rather than just the data. Experiments on ImageNet-100 and ImageNet-1k show that changing model width, input grid, stride, and architecture (e.g., ResNet vs. ViT) shifts the crossover point where different selection criteria (easy-first vs. geometric coverage) become optimal. These findings demonstrate that the relative value of a fixed subset of samples varies with the target learner’s capacity and structure, and that selection strategies must be tuned to the specific model they will train.
By Yangze Liu, Xiao-Long Yin, Zhongyi Han
The paper introduces a winner‑take‑all training strategy for flow‑matching generators that injects a random low‑dimensional conditioning code to create multiple candidate velocity predictions. By applying the loss only to the best‑matching candidate, the method allows distinct modes to shape separate predictions, mitigating the quality‑coverage tradeoff inherent in traditional flow matching. The approach also enables a new auxiliary conditioning‑scaling guidance, and demonstrates significant improvements on ImageNet‑256, reducing Fréchet Distance scores by more than 70% on both SiT and RAEv2 testbeds.
By Wenxuan Peng, Richard Zhang, Haitian Zheng, Jaskirat Singh, Tsai-Shien Chen, Eli Shechtman, Bharath Hariharan, Hadar Averbuch-Elor, Zongze Wu
The paper evaluates coreset selection methods by incorporating both selection and training time into a unified wall‑clock budget, using a standardized benchmark across four datasets and multiple selectors. Across numerous budget anchors, simple random or full‑data training consistently outperforms sophisticated selectors, and selection costs are dominated by a full‑dataset scan that cannot be amortized. The study also identifies when subset reuse can justify selection and reports several correctness fixes in a popular codebase.
By Yangze Liu, Zhongyi Han
arXiv:2608. 09091v1 Announce Type: cross Abstract: Transfer learning is particularly useful in settings with limited training data, and within image classification it is common to transfer learn upon massive datasets like ImageNet , CIFAR-100, or COCO .
By Jing Ning, James D. Braza
The paper introduces a new active learning signal for object detection that relies on a supervised contrastive term added to the training objective. This term shapes an embedding space where distance reflects class membership, allowing an unlabeled detection to be scored by its distance from the predicted category’s region weighted by confidence—all from a single forward pass of one network. Experiments on PASCAL VOC and MS‑COCO show that this criterion outperforms the standard posterior and remains competitive with ensemble‑based methods while incurring only a modest 8.3% increase in parameters.
By Licheng Zhang, Zheng Gong
arXiv:2608.21098v1 Announce Type: new
Abstract: Fusing prior knowledge with data-driven learning is attractive where data is scarce, yet no controlled account says when it helps, is redundant, or har...
By Ahmad AlMughrabi, Albert Clop, Benjamin Busam, Ricardo Marques, Petia Radeva
The paper presents a certified rule that determines whether unseen data should be incorporated into a pretrained model’s context or its weights, based on an exact law for their information value. This law quantifies the code length saved on held‑out queries as the expected bits stored about the query, adjusted by a prior repair and decoding loss, and applies to any model, architecture, dataset, or task. The authors demonstrate that the rule’s ceilings correctly certify choices across diverse models—including transformers, TabPFN, occupancy networks, and stereo networks—and that an anytime‑valid test can reliably select the better approach (e.g., TabPFN over XGBoost or GPT‑2’s cache over fine‑tuning) with controlled error rates.
By Leon Chlon, Julie Huang
The paper investigates when per‑point data‑valuation scores (e.g., Data Shapley, influence, leverage) reduce to simple class balance. It shows that a fixed top‑k selection rule fails when a point’s value depends on what else has been chosen, leading to excess risk in equal latent‑mode models. The authors propose using class balance when labels reveal learner modes and explicit coverage otherwise, demonstrating that class‑balancing per‑point scores can match or outperform generic coverage on standard classification tasks with pretrained features, while label‑free coverage can improve accuracy by up to +0.20.
By Yingjie Bai, Weiming Zhi
arXiv:2609.36426v1 Announce Type: cross
Abstract: A detector pretrained on a broad corpus is fine-tuned on a narrow domain, its in-domain accuracy improves, and it ships. We ask what happens meanwhil...
By Trung Minh Bui, Jongsul Moon, YoungOuk Kim, Jung-Hoon Hwang, Dongin Shin
arXiv:2609.10311v1 Announce Type: cross
Abstract: The lottery ticket hypothesis posits the existence of winning tickets: sparse subnetworks that, when trained in isolation from their original initial...
By Benedikt Tscheschner, Eduardo Veas, Marc Masana
Scout-Refine-Correct (SRC) is a training recipe that equips flow-matching image generation models with an inference-time refinement capability. During sampling, the model performs a scout rollout to predict a full sample, uses this prediction as a self-conditioning signal, and then takes the actual step; this process can be repeated for multiple refinement rounds. On ImageNet-256, SRC improves LightningDiT-B’s FID from ~14.5 to ~7.26 with 550 normalised flow evaluations, matching a best-of‑N baseline that relies on an oracle verifier, and further reduces FID to 3.81 when combined with best-of‑N.
By Sergey Karpukhin, Aibek Alanov
arXiv:2605.06240v2 Announce Type: replace-cross
Abstract: Forward-Forward (FF) training lets each layer learn from a local goodness criterion. In cumulative-goodness variants, later layers can inheri...
By Amirhossein Yousefiramandi