arXiv AI July 29, 2026 By Mehdi Rahimifar, Amin Darabi, Mehran Taghian Jazi, Xing Huang, Yao Wang, Zhijun Tu, Yufei Cui, Yunke Peng, Hongliang Li

Stable FP4 Training via Transposition-Invariant Block Quantization

Read the original on arXiv AI →

arXiv:2607. 24953v1 Announce Type: cross Abstract: Reducing training precision is a key lever for improving the e ciency of large language model (LLM) training, but pushing beyond FP8 to 4-bit oating point (FP4) remains challenging due to instability during optimization.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv AI

Aug 13

Pretraining large language models with MXFP4 on Native FP4 Hardware

arXiv:2605. 09825v4 Announce Type: replace-cross Abstract: Why does full-pipeline FP4 training of large language models often diverge, even when forward activations and activation gradients remain stable?

By Musa Cim, Sarthak Arora, Poovaiah Palangappa, Miro Hodak, Ravi Dwivedula, Meena Arunachalam, Mahmut Taylan Kandemir

llms efficiency

arXiv AI

Jun 9

ScaleSweep: Accurate NVFP4 Post-Training Quantization of LLMs via Block Scale Initialization

arXiv:2606. 07618v1 Announce Type: cross Abstract: NVFP4 is a recently introduced hardware-supported FP4 format that improves the fidelity of 4-bit quantization through fine-grained block scales.

By Li Lin, Xiaojun Wan

llms efficiency

arXiv AI

Jun 9

Joint Structural Pruning and Mixed-Precision Quantization for LLM Compression

arXiv:2606. 07819v1 Announce Type: new Abstract: Recently, the efficiency of Large Language Models (LLMs) deployment has become a critical concern in practical applications.

By Hoang-Loc La, Truong-Thanh Le, Amir Taherkordi, Phuong Hoai Ha

llms efficiency benchmarks

arXiv AI

Aug 11

Attn-QAT: 4-Bit Attention With Quantization-Aware Training

arXiv:2603. 00040v3 Announce Type: replace-cross Abstract: Achieving reliable 4-bit attention is a prerequisite for end-to-end FP4 computation on emerging FP4-capable GPUs, yet attention remains the main obstacle due to FP4's tiny dynamic range and attention's heavy-tailed activations.

By Peiyuan Zhang, Matthew Noto, Wenxuan Tan, Chengquan Jiang, Will Lin, Wei Zhou, Hao Zhang

llms diffusion efficiency

arXiv Machine Learning

Jul 30

Low-Precision Training of Large Language Models: Methods, Challenges, and Opportunities

arXiv:2505. 01043v2 Announce Type: replace Abstract: Large language models (LLMs) have achieved impressive performance across various domains.

By Zhiwei Hao, Jianyuan Guo, Li Shen, Yong Luo, Han Hu, Guoxia Wang, Dianhai Yu, Yonggang Wen, Dacheng Tao

llms efficiency

arXiv Machine Learning

Jun 2

Quartet II: Accurate LLM Pre-Training in NVFP4 by Improved Unbiased Gradient Estimation

arXiv:2601. 22813v2 Announce Type: replace Abstract: The NVFP4 lower-precision format, supported in hardware by NVIDIA Blackwell GPUs, promises to allow, for the first time, end-to-end fully-quantized pre-training of massive models such as LLMs.

By Andrei Panferov, Erik Schultheis, Soroush Tabesh, Dan Alistarh

llms efficiency

One email a morning, machine-written

Related stories

Pretraining large language models with MXFP4 on Native FP4 Hardware

ScaleSweep: Accurate NVFP4 Post-Training Quantization of LLMs via Block Scale Initialization

Joint Structural Pruning and Mixed-Precision Quantization for LLM Compression

Attn-QAT: 4-Bit Attention With Quantization-Aware Training

Low-Precision Training of Large Language Models: Methods, Challenges, and Opportunities

Quartet II: Accurate LLM Pre-Training in NVFP4 by Improved Unbiased Gradient Estimation