arXiv:2606. 06034v1 Announce Type: new Abstract: Matrix inversion in chunk-wise parallel linear attention is a major bottleneck for long-context modeling, particularly on NPUs, where forward-substitution-based methods exhibit limited parallelism and poor hardware utilization.
By Luoming Zhang, Yuwei Ren, Kui Zhang, Tian Liu, Lingjuan Ge, Denghao Li, Matthew Harper Langston, Yin Huang, Weiliang Will Zeng, Liang Zhang
arXiv:2607. 04422v1 Announce Type: cross Abstract: Recent NVFP4 pretraining methods mainly target transformer linear layers, leaving optimizer states, optimizer arithmetic and attention underexplored in 4-bit pipelines.
By Siyu Ding, Mingchuan Ma, Jiabo Tong, Xingrun Xing, Ziming Wang, Guoqi Li
The paper introduces 3BASiL-TM, a one‑shot post‑training method for compressing large language models using a sparse plus low‑rank decomposition. It employs a novel 3‑block ADMM algorithm to minimize layer‑wise reconstruction error and a transformer‑matching refinement that jointly optimizes sparse and low‑rank components across layers. Experiments show a >30% reduction in WikiText2 perplexity gap for LLaMA‑8B under a (2:4 sparse + 64 LR) setting and a 2.5× faster compression runtime on an A100 GPU compared to state‑of‑the‑art methods.
By Mehdi Makni, Xiang Meng, Rahul Mazumder
ResidualQuant introduces a 2‑bit residual quantization scheme for KV caches in looped Transformers, leveraging the similarity of KV states across loops. By using the final‑loop KV states as a reference and applying least‑square scaling, rotations, and loop‑wise mixed precision, the method achieves accurate quantization down to INT2. Experiments on various looped Transformer models show that ResidualQuant improves the accuracy‑memory tradeoff, retaining near BF16 accuracy while reducing KV storage by 80.7% and boosting decode throughput and batch size on an RTX 5090.
By Heejun Kim, Junyoung Lee, SangLyul Cho, Dongsu Han, Insu Han, Sehoon Kim
arXiv:2607. 18745v1 Announce Type: new Abstract: We study low-precision computation of C=AB with both factors quantized.
By Piyush Sao, Narasinga Miniskar, Pedro Valero-Lara, Keita Teranishi, Sudip Seal
arXiv:2607. 25504v1 Announce Type: cross Abstract: Fine-grained weight pruning and activation sparsification have emerged as effective approaches for reducing the compute and memory cost of inference for Transformer models.
By Bowen Wang, Chi Zhang, Diyou Shen, Renzo Andri, Navaneeth Kunhi Purayil, Luca Benini