The paper investigates how knowledge distillation from Vision Transformers to smaller CNNs can cause dimensional collapse in the student’s representation space. Using SVD and Shannon entropy, the authors show that cosine‑based distillation leads to a drastic reduction in effective rank, while adding an InfoNCE objective can double the rank but harms downstream accuracy due to signal dilution. They further demonstrate that a label‑aware contrastive objective (Supervised Contrastive distillation) can maintain or improve accuracy without unnecessary rank expansion, indicating that effective rank alone is not a reliable indicator of representation quality.
By Kabir Thayani
Compression is fundamental to intelligence. A model that can represent its training data as a short code has discovered regularities that enable generalization.
arXiv:2607. 11883v1 Announce Type: new Abstract: Compression is fundamental to intelligence.
By Shikai Qiu, Marc Finzi, Yujia Zheng, Kun Zhang, Andrew Gordon Wilson
arXiv:2607. 18088v1 Announce Type: new Abstract: Standard evaluation of many recognition systems contains distribution shift by construction, since benchmarks place disjoint conditions in the training and test splits.
By Weijia Han, Lisha Qu
GEAR is a two‑stage framework that distills tabular foundation models into lightweight MLP or tree‑based predictors for efficient CPU deployment. In the first stage, synthetic covariates are used as teacher‑query locations to train the student on soft TFM targets, expanding coverage beyond observed rows. The second stage re‑anchors the student to the target distribution using real labels and out‑of‑fold teacher predictions, preventing self‑labeling leakage and improving performance. Experiments on TALENT and TabArena show that GEAR‑distilled MLPs outperform supervised MLPs by up to 2.00 AUC points on binary tasks and 1.35 on multiclass tasks, and also outperform CatBoost, while dramatically reducing inference time and memory usage.
By Qi Qin, Jiajie Zhu, Dali Chen, Yuzhao Zhang, Jia-Xing Han, Yu Su, Peng Zhang, Ying Yan, Yifan Sun
The paper introduces an unsupervised method for selecting essential neurons in overparameterized neural networks by minimizing mapping entropy (ME), a metric that quantifies the loss of discriminatory power when neurons are discarded. ME-based selection relies solely on hidden-activation statistics and, in experiments, identifies minimal teacher-consistent representations in teacher‑student networks and coherent functional-class mappings in a non‑linear Gaussian process task. Subnetworks chosen by ME outperform random subsets of the same size, especially under strong compression, on both a Gaussian process task and translation‑augmented MNIST.
By Margherita Mele, Andrea Castagna, Roberto Menichetti, Raffaello Potestio, Alessandro Ingrosso
SCOPE-OPSD introduces a Fisher‑conditioned privileged subspace for on‑policy self‑distillation (OPSD) that projects the teacher‑student residual onto a frozen rank‑64 factor derived from residual covariance and language‑model‑head Fisher sensitivity. The method adds no extra rollouts or inference modules and, across multiple Qwen3 model checkpoints and trajectory lengths, consistently matches or surpasses pure OPSD and a matched random baseline, achieving significant gains in most settings. A cross‑fitted diagnostic shows a 4.40‑fold increase in captured privileged‑gap compared to the random orientation.
By Yunmeng Chen (Chongqing Ant Consumer Finance Co., Ltd), Kunyu Wang (Alibaba Cloud Computing Co., Ltd), Peihan Li (Chongqing Ant Consumer Finance Co., Ltd), Yi Wang (Chongqing Ant Consumer Finance Co., Ltd), Shuyin Xia (Chongqing University of Posts and Telecommunications), Yi Liu (Chongqing Ant Consumer Finance Co., Ltd), Xinyong Cheng (Alibaba Cloud Computing Co., Ltd), Dehui Wang (Alibaba Cloud Computing Co., Ltd), Xiangyong Zhai (Alibaba Cloud Computing Co., Ltd), Yanxing Liu (Chongqing Ant Consumer Finance Co., Ltd), Song Liu (Chongqing Ant Consumer Finance Co., Ltd)
arXiv:2604. 21395v3 Announce Type: replace-cross Abstract: Ordinary supervised training minimises the task loss and then stops.
By Vishal Rajput
arXiv:2608. 10709v1 Announce Type: new Abstract: Quantization-Aware Training (QAT) enables the deployment of quantized models with minimal accuracy degradation.
By HyeonJun Lee, Hyeonsik Jo, Jinwoo Chung, Jangho Kim
arXiv:2503. 08038v2 Announce Type: replace-cross Abstract: In this paper, we delve deeper into the Kullback-Leibler (KL) Divergence loss and mathematically prove that it is equivalent to the Decoupled Kullback-Leibler (DKL) Divergence loss that consists of (1) a weighted Mean Square Error (wMSE) loss and (2) a Cross-Entropy loss incorporating soft labels.
By Jiequan Cui, Beier Zhu, Qingshan Xu, Zhuotao Tian, Xiaojuan Qi, Bei Yu, Hanwang Zhang, Richang Hong
The paper argues that relying solely on zero‑shot task accuracy is insufficient for evaluating quantized large language models (LLMs) because accuracy ignores changes in the full predictive distribution. It proposes a distribution‑sensitive framework that measures fidelity loss by computing statistical distances—such as Jensen‑Shannon Divergence and Total Variation Distance—between the full‑vocabulary output distributions of a full‑precision BF16 reference and its quantized counterparts. Experiments across five foundation architectures and four reasoning benchmarks show that these divergence metrics increase with stronger quantization, revealing distributional drift that top‑1 accuracy fails to capture, and suggest that mixed‑precision Q4_K schemes can offer lower divergence than uniform Q4_0 at comparable memory usage.
By Shahzeb Qamar, Lorenz Sparrenberg, Christian Bauckhage, Baha Rababah, Carson Leung, Murat Kantarcioglu, Cuneyt Gurcan Akcora, Rafet Sifa
arXiv:2606. 07596v1 Announce Type: new Abstract: Fine-tuning often introduces spurious correlations alongside task knowledge, causing systematic failures on underrepresented groups.
By Edward Sun, Dmitrii Troitskii