The article argues that on AI‑optimised NVIDIA B300 GPUs and newer, the FP8 tensor‑core matrix operation—implemented via the CRT‑based Ozaki Scheme II—can become the primary substrate for matrix‑heavy FP64 kernels while maintaining FP64‑grade accuracy. It introduces the Tensor‑Memory Equilibrium (TME) model, a Roofline extension with four parameters, to show that FP8 can match native FP64 performance under certain intensity thresholds and tile‑fusion conditions. The study identifies two notable exceptions—large dense‑square DGEMM and 3‑D FFT—where additional hardware or software adjustments are required to reach the memory roof.
whyItMatters":"The paper demonstrates that FP8, with appropriate reconstruction and deconstruction strategies, can replace native FP64 for high‑performance computing workloads on modern GPUs, potentially reducing hardware complexity and energy consumption while preserving accuracy."
By Satoshi Matsuoka
arXiv:2610. 01889v1 Announce Type: new Abstract: Should low-precision transformer inference use stochastic rounding (SR) or round-to-nearest (RN)?
By Yohan Chatelain (Krembil Centre for Neuroinformatics, CAMH, Toronto, Canada), Pablo de Oliveira Castro (Universite Paris-Saclay, UVSQ, LI-PaRAD, Versailles, France)
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 presents a 4‑bit quantization recipe, Minima: NVFP4 W4A4, that fully quantizes all linear layers—including the Gated DeltaNet (GDN) recurrent blocks—of the 27‑billion‑parameter Qwen3.8 LLM. Across a suite of benchmarks (perplexity, MMLU‑Pro, GSM8K, AIME'25, GPQA‑Diamond, LiveCodeBench, and RULER retrieval), the quantized model matches BF16 performance within seed noise while being 17.5 GiB in size and 14–19 % faster at prefill. The authors attribute this success to four mechanisms: block‑scaling of residuals, robust gate projections, the delta‑rule recurrence’s noise‑plateau behavior, and the per‑token quantization cost’s dilution over long contexts.
By Sergii Kozyrev, Davyd Maiboroda
EFQ-Softmax is a low‑bit probability‑generation technique that replaces the traditional exp‑then‑quantize path in Transformer attention. It maps shifted attention scores directly to block‑scaled E2M1 operands using an exponent‑only scale and a single affine rule, allowing the same low‑bit representation to be used for both numerator and denominator updates. Experiments on Qwen3‑8B, Qwen3‑VL‑8B‑Instruct, and WAN2.2‑TI2V‑5B show that EFQ‑Softmax maintains or improves model quality while reducing vector‑stage latency by about 40% on the A5 vector unit.
By Haohui Han (Xi'an Jiaotong University), Yuming Wan (Huawei Technologies Co., Ltd), Hongni Wang (Shandong University of Finance and Economics), Pengcheng Xie (Huawei Technologies Co., Ltd), Xiaodong Yan (Xi'an Jiaotong University), Runqi You (Xi'an Jiaotong University), Wencong Zhang (Xi'an Jiaotong University)
arXiv:2607. 04302v1 Announce Type: cross Abstract: We present HiFA4, a post-training operator-level design that executes both QK^T and PV in FlashAttention as 4-bit HIF4 Cube GEMMs for LLM inference on Ascend NPUs, while maintaining the online softmax state in FP16.
By Hui Dong, Yanzhao Li, Jie Gao, Chunlu Li, Zhiyuan Zhang, Yupeng Sun, Zhenyuan Chen, Zhiqiang Zou
arXiv:2606. 06510v1 Announce Type: cross Abstract: Conventional HPC dogma holds that native hardware FP64 silicon is the irreducible foundation of scientific computing -- the "holy grail" of double-precision simulation.
By Satoshi Matsuoka
Low-bit attention accelerates Transformer inference by moving the $QK^\top$ and $PV$ matrix multiplications to FP8 or FP4 matrix engines. However, the softmax path often evaluates shifted-score expone...
arXiv:2607. 26515v1 Announce Type: new Abstract: We present, to our knowledge, the first end-to-end FP4 RL post-training, in which both the rollout and training policies, including their forward and backward passes, operate at 4-bit precision.
By Hei Yi Mak, Shadan Golestan, Hoang Le, Mehran Taghian Jazi, Yunke Peng, Yaoyuan Wang, Yao Wang, Junsong Wang, Tianchi Hu, Fengchen He, Guipeng Hu, Tanzila Rahman, Anandharaju Durai Raju
The paper reports a large‑scale post‑training ternarisation of the Qwen3 language model, extending a conversion pipeline from the 4B to the 8B variant. Using KOTMS rotation, E2M‑ATQ adaptive ternarisation, and GPTQ‑style error compensation, the authors achieve a 1.361× perplexity ratio across three corpora and retain 78.5% of the FP16 accuracy on zero‑shot tasks, with the 8B model outperforming the 4B by 8.9 percentage points. The study also demonstrates lossless lattice‑aware packing, producing an 8.24 GiB checkpoint that preserves perplexity, and shows that direct packed execution can reach 15.52 tokens/s in 7.35 GiB, though packed GEMV remains slower than FP16 cuBLAS.
By Anirudh Malik, M Sparsh Mehra, Poojith Devan
arXiv:2608. 10103v1 Announce Type: cross Abstract: High-performance Tensor Core kernels rely on a low-level PTX pipeline built from asynchronous data movement with cp.
By Matt J. Borowski, Blazej Osinski
The paper demonstrates that greedy decoding from large language models is not precision‑invariant: the same model, prompt, and decoding algorithm can produce different outputs when run in BF16 versus FP16 on identical hardware. Across six models (1.1B–7B parameters, four families, and 12B) and three benchmarks, 49–100 % of prompts diverge, with a single token flip often cascading into trajectory‑level divergence. The authors develop an empirical error‑propagation analysis that identifies the top‑two logit margin at the LM head as the key factor, and they propose a low‑overhead intervention—selective FP32 LM head recomputation—that improves exact agreement by 22–36 percentage points with less than 4 % latency overhead.
"whyItMatters":"The findings reveal that precision choices can fundamentally alter model outputs, challenging the assumption of deterministic greedy decoding and highlighting the need for precision‑aware inference strategies."
By Gaoyuan Du, Anam Nawaz Khan, Rex Zhou, Xiaoyang Liu, Deepayan Chakrabarti, Fnu Suya, Xueping Li