TAP-Path is a task‑adaptive compression framework that restructures a pretrained Virchow2 encoder for histopathology. It selectively removes transformer blocks, prunes patch tokens, and adds a lightweight gated task head, reducing parameters by 24.96% and FLOPs by 35.20% while maintaining high accuracy on a 32‑class benchmark. The method achieves competitive test metrics and improved rare‑class performance, with strong external validation on CPTAC samples.
By Mehedi Hasan, Ashfak Yeafi, Md Khairul Islam
arXiv:2608.22066v1 Announce Type: cross
Abstract: Attention-based multiple instance learning (ABMIL) using pathology foundation model embeddings is effective for slide-level tasks, but exhaustive inf...
By Duncan Stothers, Ren-Chin Wu, William Lotter
arXiv:2607. 03949v1 Announce Type: cross Abstract: Pixel-wise Earth-observation (EO) foundation models are now achieving state-of-the-art performance via generated spatial embeddings.
By Zhengpeng Feng, Sadiq Jaffer, Ira Shokar, Jovana Knezevic, Mark Elvers, Clement Atzberger, Robin Young, Aneesh Naik, Niall Robinson, Andrew Blake, David Coomes, Anil Madhavapeddy, Srinivasan Keshav
arXiv:2607. 22771v1 Announce Type: cross Abstract: Picking the frozen image encoder for a 3D~CT vision--language model (VLM), together with the token-compression scheme on top of it, is a search over many candidates.
By Renjie Liang
arXiv:2608. 03796v1 Announce Type: cross Abstract: Small language models are often the only option for deployment under tight latency, cost, and on-premises constraints, but they are rarely trained from scratch: a compressed model is usually recovered through knowledge distillation (KD).
By Bakbergen Ryskulov, Iker Garc\'ia-Ferrero, David Montero, David Jansen, Ali Hashemi, Jezabel R. Garcia, Antonio Tiene, Rom\'an Or\'us
arXiv:2608. 00916v1 Announce Type: cross Abstract: Modern reranking recipes---billion-scale cross-encoders, mixture-of-experts (MoE) backbones, and distillation against strong teachers---have outpaced the training infrastructure available to most academic groups.
By Zhichao Xu, Xueguang Ma, Shengyao Zhuang, Luyu Gao, Wenqian Ye, Yu Wang, Jamie Callan, Jimmy Lin
arXiv:2603. 26556v2 Announce Type: replace-cross Abstract: Converting a pretrained Transformer into a more efficient hybrid model through distillation offers a promising approach to reducing inference costs.
By Juan Gabriel Kostelec, Qinghai Guo
The paper presents a method for cross‑architecture knowledge distillation from a fine‑tuned DINOv2 Vision Transformer teacher to a lightweight bidirectional Visual State Space Model (LVSSM) student for tea leaf disease classification. By addressing training‑stability issues with a progressive convolutional stem and gated selective‑scan block, the 4.45 M‑parameter student achieves a mean test accuracy of 95.41%—a 3.09‑point improvement over the teacher’s 92.32%—while using only one‑fifth of the teacher’s parameters. Ablation studies show that simple logit‑level distillation outperforms intermediate feature alignment, and the gains are specific to students that start below the teacher’s performance.
By Zibo Zhou, Zongsen Qiu, Rui Chen, Yujie Yao, Yue Zhou, Jianjun Wang
The paper investigates whether small models distilled from larger ones behave similarly when using byte versus token tokenization. It introduces two methods—Marginalize‑It (approximate) and End‑Of‑Token (exact)—to convert token logits to byte logits, and conducts a large‑scale study on decoder‑only dense transformers ranging from 1 billion to 1 trillion bytes of data. Results show that while token‑based models excel early, byte‑based models eventually surpass them with more compute, achieving higher performance ceilings, greater data efficiency, and lower logit storage costs.
By Kalyani Marathe, Artidoro Pagnoni, Tomasz Limisiewicz, Margaret Li, Mike Lewis, Luke Zettlemoyer, Srinivasan Iyer
Video Diffusion Transformers (DiTs) spend most of their compute inside the Self-Attention operation, whose cost grows quadratically, $\mathcal{O}(n^2)$, with the number of latent tokens $n$. For the task of video generation, the token count is large, so this term dominates runtime and memory, and thereby caps the resolution and duration we can generate.
arXiv:2606. 03879v1 Announce Type: cross Abstract: As foundation models scale toward fusing more heterogeneous visual streams, understanding how diverse encoders interact under joint training becomes a prerequisite for principled design.
By Wei Ding, Yudong Zhang, Ruobing Xie, Xingwu Sun, Jiansheng Chen, Yu Wang
Dyna‑DINO introduces a curriculum for Vision Transformer (ViT) knowledge distillation that uses the teacher’s intermediate feature maps as progressively harder targets, enabling a student to build foundational representations before tackling higher‑level abstractions. The approach accelerates convergence and improves performance across multiple tasks: on ImageNet‑100 the distilled ViT‑S reaches 90.1% accuracy (+12.24% over baseline), while on ImageNet‑1K it yields +3.9% and +6.09% gains on Oxford and Paris retrieval, +1.93% on semantic segmentation, and notable classification improvements. Additionally, the curriculum reduces training FLOPs by 25.1% and training time by 21% on ImageNet‑100 through early‑stopping of teacher inference.
By Jiaqi Zhang, Ashton Lee, Anthony Wong, John Zou, Sami BuGhanem, Randall Balestriero