The paper introduces Hardware‑aware Calibrated Clustered Attention (BC attention) to speed up the global attention layers of the Visual Geometry Grounded Transformer (VGGT) used for 3D scene reconstruction. By clustering queries within hardware‑friendly neighborhood blocks and adding hashing hyperplane calibration plus threshold‑based error compensation, BC attention reduces computation and memory traffic, enabling it to handle long sequences with practical latency gains on GPUs. Experiments show that calibrated BC attention accelerates global attention by 2.10–2.63× and the entire backbone by 1.77–2.35×, with only a 1% loss in accuracy, and achieves up to 2.87× latency improvement with less than 5% performance drop.
By Weitian Wang, Shubham Rai, Cecilia De La Parra, Akash Kumar
arXiv:2606. 14757v1 Announce Type: cross Abstract: Though Vision Transformers (ViTs) have become the dominant backbone in many computer vision tasks, due to permutation equivariance, their attention mechanism lacks explicit spatial inductive biases.
By Leyla Naz Candogan, Arshia Afzal, Pol Puigdemont, Volkan Cevher
arXiv:2609.23733v1 Announce Type: new
Abstract: Feed-forward visual geometry models such as the Visual Geometry Grounded Transformer (VGGT) have recently enabled direct 3D reconstruction from multi-v...
By Abteen Arab, Guile Wu, Chengjie Huang, Dongfeng Bai
arXiv:2608. 04405v1 Announce Type: cross Abstract: Long-context large language models (LLMs) are increasingly deployed in real-world applications, yet self-attention remains a major efficiency bottleneck -- especially during decoding -- due to the necessity of repeatedly processing ever-growing key-value (KV) caches.
By Daohai Yu, Zhanpeng Zeng, Keyu Chen, Wenhao Li, Zhifeng Shen, Luxi Lin, Ruizhi Qiao, Xing Sun, Rongrong Ji
arXiv:2606. 01790v1 Announce Type: cross Abstract: Vision-language-model-based graphical user interface (GUI) agents have shown broad automation capabilities, yet deployment is bottlenecked by a key-value (KV) cache that grows linearly with interaction steps.
By Yuhang Han, Wenzheng Yang, Yujie Chen, Xiangqi Jin, Yaojie Zhang, Siteng Huang, Linfeng Zhang
The paper presents a retraining‑free technique that transforms pre‑trained transformers into elastic models capable of adapting to any computational budget on the fly. By jointly ranking every MLP neuron and attention head on a common scale and applying soft pruning across multiple sparsity targets, the method achieves smooth performance degradation up to 60% sparsity while outperforming one‑shot baselines. Experiments on 27 vision and text transformer encoders show significant parameter reductions with minimal accuracy loss and notable throughput and latency gains on GPUs, all produced in minutes.
By Walter Simoncini, Dheeraj Varghese, Gertjan J. Burghouts, Cees G. M. Snoek
HGSQ is a real‑time aerial small‑object detector that uses a heatmap‑guided sparse query strategy to focus computation on foreground regions. It introduces a lightweight Heatmap Budget Predictor to generate a foreground budget map, and then employs Heatmap‑Guided Sparse Query Selection, Heatmap‑Gated Lite Snake Convolution, and Adaptive Query‑Decoder Budgeting to efficiently process only small‑object areas. On NWPU VHR‑10 and VisDrone2019, HGSQ achieves 95.10 mAP50 and 54.8 mAP50 respectively while running at 96 FPS with only 48.6 GFLOPs on an RTX 4070.
By Yangchen Zeng
arXiv:2607.09086v2 Announce Type: replace
Abstract: We present Subtoken Vision Transformer (SubViT), a selective image tokenization method for fine-grained visual recognition. Standard Vision Transfo...
By Jie Zhu, Ivy Zhang, Minchul Kim, Xiaoming Liu
arXiv:2611.07994v1 Announce Type: new
Abstract: Tiny object detection remains challenging because tiny instances occupy only a few pixels, and their fine-grained details are easily lost during featur...
By Zehao Wang, Si Wu, Guiping Cao, Xin Li, Yaowei Wang
arXiv:2603. 26551v2 Announce Type: replace-cross Abstract: Vision backbone networks play a central role in modern computer vision.
By Moritz Nottebaum, Matteo Dunnhofer, Christian Micheloni
TriCCOT is a tri-part architecture designed for onboard space object detection that balances computational efficiency with robust performance. It combines a convolutional region proposal network, a conformal prediction stage that enlarges bounding boxes with distribution‑free probabilistic coverage, and Aper‑GATES—a hardware‑friendly attention‑based classifier that replaces standard transformer operations with convolutional projections and gating. Experiments on DIOR and VDVRaw datasets show competitive detection accuracy and improved robustness to blur and noise, and the model was fully deployed on a Xilinx Versal VCK190 FPGA without altering the underlying DPU architecture.
By Adrien Dorise, Marjorie Bellizzi, Julia Cohen, St\'ephane May
arXiv:2609.23431v1 Announce Type: new
Abstract: Human-object interaction (HOI) detection requires grounding an interacting human-object pair and recognizing the verb that links them, often under seve...
By Junwen Chen, Keiji Yanai