FAME is an FPGA-based platform that evaluates approximate multipliers directly in hardware, eliminating slow CPU/GPU LUT emulation and reducing evaluation time for DNN inference. It also introduces a pattern-guided retraining method that uses multiplier-specific patterns to recover accuracy losses. Experiments on ResNet‑18 and MobileNetV2 over ImageNet show up to 3.47× faster multiplier evaluation and a 65.5% accuracy improvement over prior retraining approaches.
By Rappy Saha, Nima Amirafshar, Jude Haris, Nima Taherinejad, Jos\'e Cano
arXiv:2607. 03652v1 Announce Type: cross Abstract: Transformer blocks are prevalent in large language model (LLM) but present deployment challenges due to their challenging computational and memory demands.
By Victor Agostinelli, Nicolas Bohm Agostini, Antonino Tumeo
arXiv:2608. 15602v1 Announce Type: cross Abstract: While binary quantization theoretically promises extreme compression and acceleration for Large Language Models (LLMs), existing research often overlooks the necessity of specialized hardware kernels, thus failing to unleash the full acceleration potential due to persistent reliance on expensive floating-point arithmetic or runtime dequantization overheads.
By Qingyao Yang, Runming Yang, He Xiao, Wendong Xu, Junyu Chen, Haobo Liu, Chenchen Ding, Ruihan Hu, Yik-Chung Wu, Ngai Wong
arXiv:2606. 31938v1 Announce Type: cross Abstract: Deploying Vision Transformer (ViT) models on edge platforms remains challenging due to their high computational demands and the architectural heterogeneity of modern hybrid ViT models, which incorporate both fully connected and convolutional layers.
By Hubert Dymarkowski, Xingjian Fu, Rappy Saha, Jude Haris, Jos\'e Cano
arXiv:2606. 02781v1 Announce Type: cross Abstract: Deep neural networks (DNNs) have achieved state-of-the-art performance across diverse domains.
By Sohan Salahuddin Mugdho, Md. Shahedul Hasan, Brahmdutta Dixit, Yang Lv, Jian-Ping Wang, Cheng Wang
HBQ: Hierarchical Scaling Block Quantization with Hardware‑Efficiency‑Aware Design for Accurate LLM Inference proposes a new block‑quantization scheme that uses large blocks and low‑overhead significand scaling to balance hardware efficiency and accuracy. The authors demonstrate that larger blocks improve efficiency by amortizing dequantization and accumulation costs, while their SIG scaling compensates for the resulting accuracy loss. Experiments on a 28 nm ASIC accelerator show that HBQ achieves up to 4.6× higher area/energy efficiency than state‑of‑the‑art weight‑only quantization, with 1.5–3.0× speedup and 1.6–3.3× system energy reduction over existing BQ methods.
By Chun-Ting Chen, Dongmin Han, Hangyeol Mun, Jake Hyun, Arnab Raha, Amit Agarwal, Mark Anders, Mohamed Abdelfattah, Jae-sun Seo
arXiv:2407. 02362v3 Announce Type: replace-cross Abstract: Modern deep neural networks heavily rely on a large number of multiply-accumulate operations, which constitute the predominant computational cost.
By Xuqi Zhu, Huaizhi Zhang, JunKyu Lee, Jiacheng Zhu, Chandrajit Pal, Sangeet Saha, Klaus D. McDonald-Maier, Xiaojun Zhai
arXiv:2512. 12850v3 Announce Type: replace-cross Abstract: Low-latency, resource-efficient neural network inference on FPGAs is essential for applications demanding real-time capability and low power.
By Duc Hoang, Aarush Gupta, Philip Harris
arXiv:2607. 28418v1 Announce Type: cross Abstract: Pruning is a promising approach for improving the efficiency of LLMs.
By Haozhe Hu, Hao Wu, Peiran Yin, Chao Han, Yunpu Ma, Xiaoyu Shen
arXiv:2506. 20686v2 Announce Type: replace-cross Abstract: Recent advances in biomolecular modeling have been catalyzed by models such as AlphaFold3 (AF3), which introduce science-informed changes to the transformer architecture.
By Hoa La, Ahan Gupta, Alex Morehead, Jianlin Cheng, Minjia Zhang
arXiv:2605. 24391v2 Announce Type: replace-cross Abstract: As the demand for deep learning grows, cost reduction through quantization has become essential for both training and inference.
By Dahoon Park, Jahyun Koo, Sangwoo Hwang, Jaeha Kung
arXiv:2608. 13426v1 Announce Type: cross Abstract: Transformer-based language models achieve strong performance but incur substantial inference cost due to repeated high-dimensional matrix multiplications.
By Zixuan Lan, Yanhong Li, Jiawei Zhou