TORQ: Two-Level Orthogonal Rotation for MXFP4 Quantization

• URL: http://arxiv.org/abs/2605.19561v1
• Date: Tue, 19 May 2026 09:05:47 GMT
• Title: TORQ: Two-Level Orthogonal Rotation for MXFP4 Quantization
• Abstract summary: We propose TORQ (Two-level Orthogonal Rotation for MXFP4 Quantization) as a training-free Post-Training Quantization framework.<n>We show that TORQ significantly improves the accuracy of MXFP4 activation quantization compared to existing methods.
• Score: 13.984796236174331
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: As Large Language Models (LLMs) advance toward practical deployment, the Microscaling FP4 (MXFP4) format has emerged as a cornerstone for next-generation low-bit inference, owing to its ability to balance high dynamic range with hardware efficiency. However, directly applying MXFP4 to LLM activation quantization inevitably leads to significant accuracy degradation. In this paper, we theoretically analyze the error structure of MXFP4 activation quantization, revealing that the root cause of this performance drop lies in two structural imbalances between activation distributions and the MXFP4 block floating-point format: (1) extreme inter-block variance imbalance and (2) intra-block codebook utilization imbalance. To address these challenges, we propose TORQ (Two-level Orthogonal Rotation for MXFP4 Quantization), a training-free Post-Training Quantization (PTQ) framework designed to reshape the geometric properties of the activation space through optimal coordinate transformations. At the macroscopic level, TORQ leverages the Schur-Horn theorem to redistribute activation energy via inter-block orthogonal rotation, preventing high-variance blocks from driving up shared scaling factors and thereby preserving the precision of small-magnitude elements. At the microscopic level, TORQ employs maximum-entropy-guided intra-block rotation to alleviate codebook collapse and maximize the MXFP4 codebook's information capacity. Experiments on mainstream LLMs such as LLaMA3 and Qwen3 show that TORQ significantly improves the accuracy of MXFP4 activation quantization compared to existing methods: on Qwen3-32B, the perplexity on WikiText is reduced to 8.43 (vs. 7.61 for BF16), and the average accuracy increases from 38.40% with direct RTN to 73.63% (vs. 74.82% for BF16), substantially narrowing the gap between 4-bit floating-point quantization and full-precision inference.

Related papers

• Stable FP4 Training via Transposition-Invariant Block Quantization [14.373963304887754]
  In conventional 1D block quantization, forward and backward passes assign di erent scaling factors to the same values after transposition.<n>We propose a low-precision training framework based on 2D block FP4 quantization, which enforces transposition-invariant scaling and preserves consistency between forward and backward computations.<n>Our approach achieves stable end-to-end FP4 training and closely matches BF16 performance, with less than 1.3% degradation in perplexity and downstream accuracy.
  arXiv  Detail & Related papers  (2026-07-27T18:03:34Z)
• QUADS: Stabilizing NVFP4 Reinforcement Learning for MoE via QUantization-error Alignment across Dual Sides [26.939321070753607]
  Rollout generation is a major bottleneck in Reinforcement Learning for Mixture-of-Experts (MoE) Large Language Models.<n>We propose QUantization-error Alignment across Dual Sides (QUADS) to stabilize NVFP4 RL for MoE.<n>QUADS achieves BF16-level accuracy, improves average pass@1 by 21.49 points over naive NVFP4 RL, and delivers 16% higher rollout throughput than FP8.
  arXiv  Detail & Related papers  (2026-07-17T10:21:27Z)
• SharQ: Bridging Activation Sparsity and FP4 Quantization for LLM Inference [4.219937026460372]
  SharQ bridges activation sparsity and FP4 quantization through an online sparse--dense decomposition.<n>A sparse FP4 GEMM processes the backbone while a dense FP4 GEMM compensates for both mask-induced activation loss and sparse-path quantization error.
  arXiv  Detail & Related papers  (2026-06-25T04:19:04Z)
• Characterizing the Impact of NVFP4 Quantization for Low-Power Edge AI Deployment [9.460818703756205]
  Energy-efficient neural-network inference at the edge requires reducing arithmetic cost, memory traffic, energy, and storage overhead while maintaining acceptable accuracy.<n>This paper presents an ablation-focused study of NVFP4 quantization for edge-efficient neural networks.
  arXiv  Detail & Related papers  (2026-06-03T03:09:59Z)
• HARP: Hadamard-Preconditioned Adaptive Rotation Processor for Extreme LLM Quantization [48.87937677659571]
  We introduce HARP (Hadamard-preconditioned Adaptive Rotation Processor), a learnable structured two-sided processor that replaces fixed Hadamard mixing.<n>Across 2-4 bit settings on models ranging from 1B to 70B parameters, HARP improves perplexity and zero-shot accuracy over fixed RHT.<n>HarP preserves deployment efficiency, reaching 128 tok/s versus 61 tok/s for FP16.
  arXiv  Detail & Related papers  (2026-05-28T12:24:15Z)
• DiRotQ: Rotation-Aware Quantization for 4-bit Diffusion Transformers [3.0583214514538084]
  Diffusion Transformers (DiTs) achieve state-of-the-art image generation quality but incur substantial memory and computational costs at inference.<n>Existing approaches, including smoothing-based methods, mixed-precision schemes, rotation techniques, and low-rank residual methods, partially mitigate this issue but still leave a noticeable gap to FP16/BF16 performance.<n>We introduce DiRotQ, a W4A4 PTQ framework that mitigates this degradation through rotation-aware activation quantization.
  arXiv  Detail & Related papers  (2026-05-16T00:52:00Z)
• Multi-Scale Dequant: Eliminating Dequantization Bottleneck via Activation Decomposition for Efficient LLM Inference [12.249074183271743]
  Quantization is essential for efficient large language model (LLM) inference.<n>Dequantization step-converting low-bit weights back to high-precision for matrix multiplication has become a critical bottleneck on modern AI accelerators.<n>This paper presents Multi-Scale Dequant (MSD), a quantization framework that removes weight/KV dequantization from the GEMM critical path.
  arXiv  Detail & Related papers  (2026-05-13T09:49:56Z)
• Pretraining large language models with MXFP4 on Native FP4 Hardware [6.139566055770847]
  We study why full-pipeline FP4 training of large language models often diverges, even when forward activations and activation gradients remain stable.<n>We find that FP4 training instability is driven by structured micro-scaling errors along sensitive gradient paths, rather than by insufficient Instinctity.
  arXiv  Detail & Related papers  (2026-05-11T00:04:07Z)
• BPDQ: Bit-Plane Decomposition Quantization on a Variable Grid for Large Language Models [56.504879072674015]
  We propose Bit-Plane Decomposition Quantization (BPDQ), which constructs a variable quantization grid via bit-planes and scalar coefficients.<n>BPDQ enables serving Qwen2.5-72B on a single GTX 3090 with 83.85% GSM8K accuracy (vs. 90.83% at 16-bit)
  arXiv  Detail & Related papers  (2026-02-04T02:54:37Z)
• ARCQuant: Boosting NVFP4 Quantization with Augmented Residual Channels for LLMs [4.431548809730958]
  ARCQuant is a framework that boosts NVFP4 performance via Augmented Residual Channels.<n>We show that ARCQuant achieves state-of-the-art accuracy, comparable to full-precision baselines in perplexity and downstream tasks.
  arXiv  Detail & Related papers  (2026-01-12T12:27:22Z)
• Bridging the Gap Between Promise and Performance for Microscaling FP4 Quantization [77.67818998672516]
  We present the first comprehensive study of MXFP4 and NVFP4 for post-training quantization.<n>We introduce Micro-Rotated-GPTQ (MR-GPTQ), a variant of the classic GPTQ quantization algorithm.<n>We show that MR-GPTQ matches or outperforms state-of-the-art accuracy.
  arXiv  Detail & Related papers  (2025-09-27T09:22:21Z)
• Optimizing Large Language Model Training Using FP4 Quantization [73.55459961002371]
  Quantized training presents a promising solution by enabling low-bit arithmetic operations to reduce costs.<n>This work introduces the first FP4 training framework for large language models (LLMs)
  arXiv  Detail & Related papers  (2025-01-28T18:04:50Z)
• AMXFP4: Taming Activation Outliers with Asymmetric Microscaling Floating-Point for 4-bit LLM Inference [6.699442219974261]
  AMXFP4 is a 4-bit asymmetric FP format that handles both issues using asymmetric shared scales.<n>AMXFP4 outperforms MXFP4 by 3% on VQA and exceeds rotation-based methods by 1.6% on CSQA.
  arXiv  Detail & Related papers  (2024-11-15T03:11:19Z)
• AffineQuant: Affine Transformation Quantization for Large Language Models [58.45460102764]
  Post-Training Quantization (PTQ) has emerged as a subject of considerable interest due to its compression efficiency and cost-effectiveness in the context of training. Existing PTQ methods for Large-scale Language Models (LLMs) limit the optimization scope to scaling transformations between pre- and post-quantization weights. In this paper, we advocate for the direct optimization using equivalent Affine transformations in PTQ (AffineQuant)
  arXiv  Detail & Related papers  (2024-03-19T08:40:21Z)
• CBQ: Cross-Block Quantization for Large Language Models [66.82132832702895]
  Post-training quantization (PTQ) has played a key role in compressing large language models (LLMs) with ultra-low costs.<n>We propose CBQ, a cross-block reconstruction-based PTQ method for LLMs.<n> CBQ employs a cross-block dependency using a reconstruction scheme, establishing long-range dependencies across multiple blocks to minimize error accumulation.
  arXiv  Detail & Related papers  (2023-12-13T07:56:27Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.