HARP: Hadamard-Preconditioned Adaptive Rotation Processor for Extreme LLM Quantization

• URL: http://arxiv.org/abs/2605.29843v1
• Date: Thu, 28 May 2026 12:24:15 GMT
• Title: HARP: Hadamard-Preconditioned Adaptive Rotation Processor for Extreme LLM Quantization
• Abstract summary: 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.
• Score: 48.87937677659571
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Post-training quantization (PTQ) is essential for deploying LLMs under memory and bandwidth constraints. However, extreme low-bit quantization remains highly sensitive to activation outliers and anisotropic weight curvature. Existing incoherence-based PTQ methods mitigate this issue with fixed randomized Hadamard transforms (RHTs), which improve quantization robustness but cannot adapt the rotated basis to the layer, calibration distribution, or quantizer. We introduce HARP (Hadamard-preconditioned Adaptive Rotation Processor), a learnable structured two-sided orthogonal processor that replaces fixed Hadamard mixing while preserving exact full-precision equivalence. HARP represents each rotation as a product of sparse butterfly-like block-orthogonal stages, supports non-power-of-two dimensions via Mixed-Radix schedules, and initializes to the RHT processor up to a fixed permutation. Fitted only on calibration data, HARP adapts the quantization basis to each layer and backend. Across 2-4 bit settings on models ranging from 1B to 70B parameters, HARP improves perplexity and zero-shot accuracy over fixed RHT. Importantly, HARP preserves deployment efficiency, reaching 128 tok/s versus 61 tok/s for FP16.

Related papers

• COEC: Calibrated Orthogonal-Equivalence Compensation for Structured Pruning of Large Language Models [13.501861808796873]
  We propose COEC (Calibrated Orthogonal-Equivalence Compensation) as a training-free compensation framework for structured pruning.<n>Experiments on the Llama-3, Llama-3.1, and Qwen2.5 model families show that COEC improves perplexity on every model and zero-shot accuracy in most settings.
  arXiv  Detail & Related papers  (2026-08-21T14:20:49Z)
• Closed-Form Spectral Regularization for Multi-Task Model Merging [96.82449201305234]
  Model merging combines several independently fine-tuned experts into a single multi-task model without any training data.<n>State-of-the-art merging methods formulate merging as a layer-wise interference problem.<n>We propose SWUDI, a closed-form method that combines a soft exponential filter, which matches the gradient-flow trajectory of iterative descent.
  arXiv  Detail & Related papers  (2026-06-05T14:00:47Z)
• Minimizing the Hidden Cost of Scales: Graph-Guided Ultra-Low-Bit Quantization for Large Language Models [50.16014098038291]
  Post-training quantization (PTQ) is critical for the efficient deployment of large language models (LLMs)<n>We propose SAGE-PTQ, a novel ultra-low-bit quantization framework for LLMs that minimizes hidden scaling cost.<n>On LLaMA-3B, SAGE-PTQ achieves 6.74 WikiText2 perplexity, compared to 55.8 for BiLLM, while using less than 50% of BiLLM's GPU memory.
  arXiv  Detail & Related papers  (2026-06-03T20:51:52Z)
• 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)
• Reframing preprocessing selection as model-internal calibration in near-infrared spectroscopy: A large-scale benchmark of operator-adaptive PLS and Ridge models [0.2609784101826761]
  This paper studies the case where that search can be collapsed into one calibration step.<n> Linear operator-adaptive calibration gives comparable prediction quality to exhaustive preprocessing screening.
  arXiv  Detail & Related papers  (2026-05-13T14:23:00Z)
• ReSpinQuant: Efficient Layer-Wise LLM Quantization via Subspace Residual Rotation Approximation [32.6734400537211]
  Post-Training Quantization (PTQ) has emerged as a promising solution for mitigating activation outliers in the quantization of Large Language Models (LLMs)<n>Layer-wise transformation methods emerged, achieving superior accuracy through localized adaptation.<n>However, layer-wise methods cannot fuse activation rotation matrices into weights, requiring online computations and causing significant overhead.<n>We propose ReSpinQuant, a quantization framework that resolves such overhead by leveraging offline activation rotation fusion and matching basis.
  arXiv  Detail & Related papers  (2026-04-13T07:00:26Z)
• SERQ: Saliency-Aware Low-Rank Error Reconstruction for LLM Quantization [7.372706701787234]
  Post-training quantization (PTQ) has emerged as a prevailing technique for deploying large language models efficiently.<n>We propose SERQ, a saliency-aware error reconstruction method for low-bit LLM inference that employs a single low-rank compensation matrix.
  arXiv  Detail & Related papers  (2026-03-09T10:04:12Z)
• 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)
• LRQ-DiT: Log-Rotation Post-Training Quantization of Diffusion Transformers for Image and Video Generation [41.66473889057111]
  Diffusion Transformers (DiTs) have achieved impressive performance in text-to-image and text-to-video generation.<n>DiTs' high computational cost and large parameter sizes pose significant challenges for usage in resource-constrained scenarios.<n>We propose LRQ-DiT, an efficient and accurate post-training quantization framework for image and video generation.
  arXiv  Detail & Related papers  (2025-08-05T14:16:11Z)
• BASE-Q: Bias and Asymmetric Scaling Enhanced Rotational Quantization for Large Language Models [25.783531928577233]
  BASE-Q is a simple yet powerful approach that combines bias correction and asymmetric scaling to reduce rounding and clipping errors.<n>Experiments demonstrate the effectiveness of BASE-Q, narrowing the accuracy gap to full-precision models by 50.5%, 42.9%, and 29.2% compared to QuaRot, SpinQuant, and OSTQuant, respectively.
  arXiv  Detail & Related papers  (2025-05-26T14:22:21Z)
• FlatQuant: Flatness Matters for LLM Quantization [58.28221892035609]
  We propose FlatQuant, a new post-training quantization approach that enhances the flatness of weights and activations.<n>Our approach identifies optimal affine transformations for each linear layer, calibrated in hours via a lightweight objective.<n>It achieves less than 1% accuracy drop for W4A4 quantization on the LLaMA-3-70B model, surpassing SpinQuant by 7.5%.
  arXiv  Detail & Related papers  (2024-10-12T08:10:28Z)
• 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)
• PAMS: Quantized Super-Resolution via Parameterized Max Scale [84.55675222525608]
  Deep convolutional neural networks (DCNNs) have shown dominant performance in the task of super-resolution (SR) We propose a new quantization scheme termed PArameterized Max Scale (PAMS), which applies the trainable truncated parameter to explore the upper bound of the quantization range adaptively. Experiments demonstrate that the proposed PAMS scheme can well compress and accelerate the existing SR models such as EDSR and RDN.
  arXiv  Detail & Related papers  (2020-11-09T06:16:05Z)

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.