Neural Network Quantization

Follow this topic Ask about this topic

Momentum

57 papers in the last four weeks, up 54% on the four weeks before. 0.4% of all new papers.

Jul 13Week of Sep 28

Latest papers 312

Jul 5, 2026cs.LG

HiFA4: Training-Free 4-bit FlashAttention on Ascend HIF4 NPUs for LLM Inference

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. To our knowledge, HiFA4 is the first Ascend-HIF4-targeted design of this kind evaluated on standard NLP benchmarks. HiFA4 combines two mechanisms. Smooth-QK applies a calibration-static per-channel equivalent rescaling to Q and K after RoPE, transferring quantization difficulty from K to Q without per-tile online reduction at inference. P-Reordering accumulates the softmax normalizer from the same quantized attention weights P_hat used in the PV GEMM, rather than from a higher-precision reconstruction. We show that this inconsistent formulation introduces a coherent output-scaling error, and validate the effect on a Qwen3-8B Layer-0 MMLU trace, where all 3.6M measured attention tiles exhibit net probability-mass loss with median epsilon_bar = -0.064. P-Reordering also allows the normalizer to be fused into the PV Cube GEMM. Across five LLMs, HiFA4 reduces quantization-induced decision drift. On Qwen3-8B, it recovers 37.5% of the accuracy gap introduced by direct HIF4 quantization, narrows the sample-weighted accuracy loss from 1.12 pp to 0.70 pp, reduces BF16-inconsistent MMLU predictions from 16.3% to 8.2%, and cuts MMLU accuracy regressions by 57% (1071 to 465). On Gemma2-9B, mild smoothing keeps HiFA4 within 0.7 pp of BF16 while reducing MMLU regressions by 27%. On LLaMA3.1-8B, Mistral-7B, and Phi-4B, where Smooth-QK is disabled, P-Reordering with the adopted Q-Mean auxiliary still reduces full-set MMLU regressions by 41-52%. A preliminary instruction-scheduling analysis projects a 35.4% critical-path latency reduction relative to BF16 by fusing the softmax normalizer into the PV Cube GEMM; on-hardware validation is left to future work.

Hui Dong, Yanzhao Li, Jie Gao +5

Huawei Technologies

Softmax Attention LLM Quantization AI Accelerator Inference 4-Bit Quantization Attention Quantization Post-Training Quantization+1

Jul 4, 2026cs.LG

A Unified Framework for Quantized and Continuous Strong Lottery Tickets

The Strong Lottery Ticket Hypothesis (SLTH) asserts that sufficiently overparameterized, randomly initialized neural networks contain sparse subnetworks that, even without any training, can match the performance of a small trained network on a given dataset. A key mathematical tool in the theoretical study of SLTH has been the Random Subset Sum Problem (RSSP). The SLTH has recently been extended to the quantized setting, where the network weights are sampled from a discrete set rather than from a continuous interval. These new results are however far from those in arbitrary-precision setting in several ways. In this work, we provide an analysis of the RSSP in the discrete setting, and use it to derive tight SLTH guarantees in the quantized case. Our analysis obtain tight bounds on the failure probability of finding a strong lottery ticket in the quantized regime, providing an exponential improvement over previous results. Most importantly, it unifies the literature by showing that both approximate representations in the continuous setting and exact representations in quantized settings naturally emerge as limiting cases of our results. This perspective not only sharpens existing bounds but also provides a cohesive framework that simultaneously handles approximation and rounding errors.

Aakash Kumar, Emanuele Natale

COATI, CNRS, Inria, I3S, Université Coté d’Azur, France

Neural Network Approximation Theory Neural Network Quantization

Jul 4, 2026cs.AR

ELiTeFormer: An Efficient Transformer for FPGAs

Transformer blocks are prevalent in large language model (LLM) but present deployment challenges due to their challenging computational and memory demands. While prior work has typically optimized attention mechanisms or feed-forward networks (FFNs) separately, few hardware (HW) architecture have jointly addressed both components with co-designed hardware acceleration. We present ELiTeFormer (Efficient Linear Ternary Transformer), the first Transformer model architecture that unifies hybrid linear attention with ultra-low-precision (ternary) linear projections, specifically co-designed for field-programmable gate array (FPGA) deployment. ELiTeFormer achieves 10x model weight compression and 12.8x key-value (KV) cache compression compared to LLaMA 3, while maintaining competitive accuracy (31.9% on the MMLU benchmark, within 3.0% of BitNet b1.58). Our key architectural contribution is a novel processing element (PE) micro-architecture that eliminates all multiplications in ternary linear projections through bitmasking operations, significantly reducing resource utilization by completely avoiding dedicated digital signal processing (DSP) blocks. We simulate, synthesize, and deploy ELiTeFormer targeting a Xilinx VCK5000 Versal board using high-level synthesis (HLS) flows. Block-level simulations show 9.6x speedup for FFN operations and 4.4x speedup for attention compared to standard implementations. End-to-end deployment achieves up to 3.9x lower latency and 3.2x better energy efficiency than LLaMA 3 on an NVIDIA A100 graphics processing unit (GPU) at long context lengths. This represents the first FPGA realization combining linear attention with ternary quantization, demonstrating the viability of algorithm-architecture co-design for next-generation LLM acceleration.

Victor Agostinelli, Nicolas Bohm Agostini, Antonino Tumeo

Oregon State University Corvallis, Oregon, USA · Pacific Northwest National Laboratory Richland, Washington, USA

LLM Quantization Ternary Quantization Transformer Inference LLM Inference Acceleration Linear Attention FPGA-Based Neural Network Acceleration+1

Jul 3, 2026cs.LG

Variable Bit-width Quantization: Learning Per-Group Precision for "Bigger-but-Smaller" Language Models

Low-bit quantization shrinks language models but treats precision as a single global hyper-parameter: every weight uses the same bit-width. We introduce Variable Bit-width Quantization (VBQ), a training-time method in which each contiguous group of 64 weights learns its own resolution from {1,2,4,8} bits via a Gumbel-Softmax relaxation, trained jointly by an alternating optimization that gives the precision logits a clean, task-aligned signal. VBQ discovers a consistent, strongly heterogeneous allocation within individual projection types, not merely across layers, impossible to express with per-layer methods: 69% of groups collapse to 1 bit, the LM head averages 1.09 bits, while the first MLP block keeps ~2.5 bits. This pattern is stable enough to freeze into a fixed recipe and reuse without further search. The recipe yields a "bigger-but-smaller" regime: a 131M model at 1.82 mean bits reaches perplexity 4.2 on TinyStories, beating a 55M FP16 model (PPL 4.4) at 3.8x less storage, and lets a 1.46B model on FineWeb-Edu match a 593M FP16 control at ~3.7x less storage with 2.5x more parameters. As quality-per-byte, VBQ is 3.9-8.4x more efficient than FP16. The recipe maps directly to packed low-bit storage, so it also accelerates inference: with custom fused dequantize-and-multiply kernels, memory-bandwidth-bound autoregressive decode is faster at equal output, and the speedup grows with scale (parity at 131M, 1.9x at 1.0B, 4.7x at 9B on Apple silicon). A distributional analysis (KL divergence and argmax-flip rate) reveals a striking mechanism: deeper layers progressively self-heal the quantization error injected by early layers. The win is a from-scratch, train-time phenomenon; scaling the search economically beyond 1.5B parameters remains open. VBQ reframes precision as a learnable, non-uniform resource and shows that spending a fixed bit budget unevenly beats spending it uniformly.

Hamish Ogilvy

Algolia Research

Mixed-Precision Quantization LLM Quantization LLM Inference Acceleration Low-Bit Quantization Neural Network Quantization

Jul 2, 2026cs.CV

OrbitQuant: Data-Agnostic Quantization for Image and Video Diffusion Transformers

Diffusion transformers (DiTs) achieve state-of-the-art image and video generation, but their multi-step sampling and growing parameter count make inference expensive. Post-training quantization (PTQ) is the natural remedy, yet DiT activations shift across timesteps, prompts, and guidance branches, forcing prior methods to re-fit calibration data for every new checkpoint or modality. We present OrbitQuant, a data-agnostic weight-activation quantizer that bypasses range estimation by quantizing in a normalized, rotated basis. In this basis, a randomized permuted block-Hadamard (RPBH) rotation concentrates each coordinate around one fixed, known marginal regardless of the input, so a single Lloyd-Max codebook serves all timesteps, prompts, and layers of a given input dimension. We extend the same quantizer to weight rows offline, absorbing the rotation into the weights so that it cancels inside each linear layer and only a forward rotation on the activations remains at runtime. The same recipe transfers from image to video with no per-modality tuning. Across FLUX.1, Z-Image-Turbo, Wan 2.1, and CogVideoX, it sets the state of the art for PTQ at several low-bit settings. It also pushes PTQ of image diffusion transformers to W2A4 with usable generation quality.

Donghyun Lee, Jitesh Chavan, Duy Nguyen +5

Cantina Labs · University of Southern California · *Work done during an internship at Cantina Labs. +1

Video Diffusion Models Diffusion Transformer Diffusion Model Quantization Rotation-Based Quantization Activation Quantization Post-Training Quantization+1

Jul 1, 2026math.OC

Boundary-Aware Quantization: Finite-Scale Decision Geometry of Neural Classifiers

We measured quantization-induced decision-boundary changes using local logit-margin radii, first-order boundary displacement, normal variation, slice-boundary Jaccard distance, grid prediction changes, multiclass junction counts, and low-margin boundary-band flips. On the digits benchmark, 8-bit weight quantization preserved all test labels while producing boundary-mask Jaccard

0.428

on the PCA slice; at 4 bits, accuracy remained

0.9733

, while boundary Jaccard rose to

0.970

and median local boundary shift reached

0.0290

. Interpolation between adjacent quantization levels localized the visible reconfigurations at multiclass junctions, with 12, 34, and 17 triple-junction cells in the selected transitions. Calibration-to-test stopping reduced the digits held-out flip rate from

0.0094

to

0.0022

and boundary Jaccard from

0.825

to

0.524

; the same stopping rule also reduced flips on MNIST and Fashion-MNIST. On official CIFAR-10 subsets, PTQ-W selected by accuracy gave 6-bit flip

0.0367

and boundary Jaccard

0.184

, whereas boundary-aware stopping selected 8-bit flip

0.0083

and boundary Jaccard

0.048

. On full CIFAR-10 with three seeds, 6-bit PTQ-W lost

0.0029

accuracy relative to float, changed

5.3\%

of held-out decisions, and changed

24.5\%

of low-margin boundary-band decisions. A fixed-bit boundary-gap rounding term changed the trade-off at 4 bits by reducing boundary Jaccard from

0.457

to

0.435

and boundary-band pair-order flip from

0.3600

to

0.3558

, with an accuracy trade-off; the 3-bit stress test exposed the tuning limit of this surrogate. Calibration boundary Jaccard predicted held-out boundary Jaccard across PTQ-W and optimized rounding variants with

r=0.947

--

0.994

.

O. M. Kiselev

Innopolis University, Innopolis, Republic of Tatarstan, Russian Federation

Weight-Only Quantization Post-Training Quantization Neural Network Quantization

Jul 1, 2026cs.CL

$\text{Log}_\text{b}$ Quant: Quantizing Language Models in Logarithmic Space

Quantization has become an invaluable tool to reduce memory requirements and inference speed of modern language models, in particular to make them available for consumer setups and edge devices. While previous work has primarily focused on uniform quantization codebooks, such approaches are prone to suboptimal representations due to low-frequency high-magnitude weights. We introduce Log

_\text{b}

Quant, a novel logarithmic quantization approach with adjustable bases, to adapt to common parameter distributions. We show that our method exhibits superior performance at 4-bit precision on several performance benchmarks compared to asymmetric linear quantization at tensor-wise granularity, while achieving moderate speedup and high memory savings, making it suitable for private use on consumer-grade GPUs.

Jeremias Bohn, Tizian Dippold, Mahdi Koubaa +2

School of Computation, Information and Technology, Technical University of Munich, Munich, Germany

LLM Quantization 4-Bit Quantization Low-Bit Quantization Neural Network Quantization

Jul 1, 2026cs.LG

Beyond Activation Alignment:The Alignment-Diversity Tradeoff in Task-Aware LLM Quantization

Mixed-precision quantization (MPQ) has become a key technique for deploying large language models under stringent memory and compute constraints. We first identify a phenomenon that we term the Perplexity Illusion: layers ranked as important by perplexity-based sensitivity show little rank correlation with those that are most influential for complex reasoning performance, with Kendall

τ\approx 0

in our analysis. We further reveal an Alignment-Diversity Tradeoff: using only target-task calibration data can degrade post-quantization performance, whereas incorporating general-domain data stabilizes sensitivity estimation and improves robustness across tasks. Based on these observations, we propose TASA (Task-Aware Sensitivity Analysis), a two-level framework that jointly optimizes calibration-data composition and mixed-precision bit allocation. Specifically, TASA searches for a calibration-data mixture using a training-free gradient-trace alignment criterion, and then aggregates perplexity and reasoning-oriented sensitivity signals to guide both inter-layer and intra-layer bit allocation. Experiments on LLaMA-3-8B and Qwen2.5-7B reveal a precision inversion: appropriately allocated 3.5-bit models can match or surpass less task-aware 4-bit baselines. At an average precision of 3.5 bits, TASA matches or outperforms several competitive 4-bit uniform baselines in aggregate accuracy, and improves over the strongest W3 baseline on GSM8K by more than 20 absolute points on LLaMA-3-8B. These results show that calibration-data composition substantially affects task-sensitive quantization, a factor underexplored in prior work.

Fei Wang, Chao Xue, Taoran Liu +3

South China University of Technology · Ocean University of China · Shenzhen Campus of Sun Yat-sen University +1

Mixed-Precision Quantization LLM Quantization Language Model Calibration Post-Training Quantization Neural Network Quantization

Jun 30, 2026cs.LG

Zero-Shot Quantization for Object Detectors using Off-the-Shelf Generative Models

With an increasing number of Object Detection (OD) models being deployed on edge devices, Zero-Shot Quantization for OD (ZSQ-OD) aims to quantize these models when access to the original training data is prohibited. Existing research on Zero-Shot Quantization-Aware Training (QAT) for OD synthesizes training sets through noise optimization. However, this approach struggles to maintain performance in low-bit regions. In this paper, we introduce GoodQ (Generative off-the-shelf models for object detector Quantization), a QAT pipeline that utilizes off-the-shelf generative models to construct a training set. We first identify three challenges that arise when introducing a generative model to the ZSQ-OD task: 1) each image contains dense information with multiple instances, 2) the class-wise distribution in the original dataset is imbalanced, and 3) the pseudo-labels assigned to the generated images can potentially act as noisy signals during QAT. GoodQ addresses these challenges by 1) introducing an Information-Dense Prompting strategy to generate multi-instance images, 2) applying Intrinsic Distribution-Aware Selection to match the pretrained class distribution, and 3) employing Teacher-guided Adaptive Noise Reduction to mitigate noise arising from the QAT process. Our framework achieves state-of-the-art performance in low-bit ZSQ (W4A4) and extends quantization to extreme bit-widths (W3A3). Furthermore, we conduct an extensive analysis to uncover the underlying factors contributing to the efficacy of GoodQ.

Hyunho Lee, Kyomin Hwang, Hyeonjin Kim +3

Seoul National University, Seoul, South Korea · LG Electronics, Seoul, South Korea

Quantization-Aware Training Synthetic Data Generation Object Detection Low-Bit Quantization Neural Network Quantization

Jun 28, 2026cs.CV

W4A4 Quantization for Inference on Wan2.2-I2V-A14B

We summarize our submission to Sub-Challenge 1: W4A4 Quantization for Inference (HiF4 / MXFP4) of the ICME 2026 Low-Bit-width Large-Model Quantization Challenge. The sub-challenge targets 4-bit weight and 4-bit activation inference on Wan-AI/Wan2.2-I2V-A14B under HiF4 or MXFP4 numerical formats. We adapt two complementary ideas from LLM quantization, MixQ-style mixed precision for sparse activation outliers and SmoothQuant-style per-channel smoothing, together with block-wise HiF4 packing for Wan2.2 feed-forward linear layers. Calibration on representative OpenS2V-5M batches identifies heavy-tailed activation channels; smoothing rebalances dynamic range before W4A4 rounding; and a dual-branch GEMM preserves outlier columns in higher precision while the bulk of channels use strict W4A4. On official VBench I2V metrics, our pipeline stays within 2-3.5 percent of FP16 on most quality axes and improves motion smoothness, outperforming a native HiFloat4 baseline that degrades roughly 5 percent relative to FP16 across all reported scores.

Yidong Chen, Chengyu Shi, Jiahao Liu

Tsinghua University

Diffusion Model Quantization Post-Training Quantization Low-Bit Quantization Neural Network Quantization

Jun 26, 2026cs.AR

SEADA: An efficient methodology for optimizing mixed-precision DNNs on multi-precision spatial architectures

Mixed-precision computation has been introduced in deep neural networks (DNNs) as an effective approach to reduce latency, energy consumption, and memory footprint. However, efficiently mapping mixed-precision networks onto multi-precision spatial architectures poses several challenges. These include determining the appropriate precision for each layer, balancing layer-wise accuracy sensitivity to quantization against architectural heterogeneity and system-level constraints, and accurately estimating the system-level cost of heterogeneous precision assignments. This work presents SEADA, an efficient methodology designed to address these challenges. SEADA comprises: (i) a configurable system-level analytical cost model of a multi-precision spatial accelerator architecture; (ii) a fast mapping tool that identifies near-optimal mappings of DNN workloads onto the target integer accelerator; (iii) analytical models for floating-point layers to estimate the overall benefits of mixed-precision execution; and (iv) a per-layer precision selection methodology based on bit-level entropy, enabling efficient assignment across multiple numerical precisions. SEADA's efficiency provides designers with a robust framework for the design-space exploration of multi-precision architectures.

Leandro Fiorin, Marco Ronzani, Cristina Silvano

Indipendent Researcher Groningen, The Netherlands · Dipartimento di Elettronica e Informazione, Politecnico di Milano. Milano, Italy

Mixed-Precision Quantization Efficient Neural Network Inference AI Accelerator Inference Neural Network Quantization

Jun 26, 2026cs.CV

Learning 1-Bit LiDAR-based Localization with Auxiliary Objective

6-DoF LiDAR-based localization is a fundamental capability for autonomous systems operating in large-scale outdoor environments. Many deep-learning-based localization methods have achieved promising performance so far. However, as one of the always-on modules competing for limited on-board computational resources, the localization module is expected to consume only a small portion of the overall compute budget. Most existing learning-based methods are still too heavy for this purpose. In contrast, binary neural networks (BNNs) offer an appealing solution, but the 1-bit compression causes severe information loss and performance drop. In this paper, we address this challenge by proposing Binarized LiDAR-based Localization (BiLoc), the first binary neural network framework for 6-DoF LiDAR localization. Specifically, we reinterpret the training of BNNs from the perspective of the information-bottleneck principle, aiming at retaining minimal yet sufficient representations for pose estimation while suppressing redundant variations. And we introduce an auxiliary objective that adaptively regulates information retention in the binary encoder, effectively mitigating the information loss caused by binarization. This auxiliary objective provides additional optimization signals that compensate for the limited representational capacity and the gradient mismatch inherent in BNNs. Extensive experiments on large-scale outdoor LiDAR datasets demonstrate that BiLoc establishes a new state of the art for LiDAR localization with BNNs.

Kaijie Yin, Zhiyuan Zhang, Tian Gao +3

University of Macau · Singapore Management University · Eastern Institute of Technology, Ningbo

Binary Neural Networks LiDAR Localization Neural Network Quantization

Jun 25, 2026cs.LG

Quantization in Federated Learning: Methods, Challenges and Future Directions

Federated Learning (FL) has become a foundational paradigm for privacy-preserving distributed intelligence, yet its scalability remains fundamentally constrained by communication bottlenecks, device heterogeneity, and the challenges of training under statistically non-IID data. Quantization is one of the most effective mechanisms for mitigating these limitations, reducing both uplink/downlink payloads and on-device computation. This paper provides the first FL-centric systematic review of quantization, introducing a novel taxonomy organized around FL-specific dimensions, including client heterogeneity, aggregation consistency, communication-scheduling adaptation, non-IID robustness, privacy/security integration, and hardware/energy co-optimization. Beyond cataloging existing methods, we analyze how quantization interacts with core FL behaviors such as client drift, partial participation, convergence stability, secure aggregation, and differential privacy. We further identify cross-method insights, open research gaps, and design guidelines for practitioners deploying quantized FL on mobile, IoT, and edge platforms. This survey thus establishes quantization not merely as a compression technique, but as a fundamental systems component shaping the performance, robustness, and practicality of modern FL.

Farwa Ikram, Dipanwita Thakur, Antonella Guzzo +1

DIMES - University of Calabria, Italy

Neural Network Quantization Federated Learning

Jun 25, 2026cs.CL

CAT-Q: Cost-efficient and Accurate Ternary Quantization for LLMs

In this paper, we present CAT-Q, Cost-efficient and Accurate Ternary Quantization, for compressing and accelerating LLMs. Unlike existing state-of-the-art ternary quantization methods that rely on data-intensive and costly quantization-aware training to mitigate severe performance degradation, CAT-Q is a simple yet effective post-training quantization scheme that is readily applicable to LLMs with diverse architectures and model sizes. It has two key components, learnable modulation (LM) and softened ternarization (ST), which are coupled from an optimization perspective. LM leverages a composition of learnable factors to modulate the distribution of pre-trained high-precision weights and the ternary threshold, making them less sensitive to ternarization. ST further introduces a differentiable transition function to guide the ternarization process toward stable convergence. We show that, for pre-trained LLMs with 1.7B to 8B parameters, CAT-Q can efficiently quantize them into ternary models using only 512 calibration samples, while achieving superior performance than the seminal BitNet 1.58-bit v1 and v2 families (with 1.3B to 7B parameters) trained with 100B tokens, yielding about a 100,000X reduction in training tokens. Moreover, we show for the first time that CAT-Q can quantize much larger pre-trained LLMs having 14B to 235B parameters into leading ternary models within just 8 to 60 hours on 8 A100-80GB GPUs. Code is available at https://github.com/IntelChina-AI/BitTern.

Shigeng Wang, Chao Li, Yangyuxuan Kang +2

1Intel Labs China

LLM Quantization Ternary Quantization LLM Compression LLM Inference Acceleration Post-Training Quantization Neural Network Quantization

Jun 25, 2026cs.LG

SharQ: Bridging Activation Sparsity and FP4 Quantization for LLM Inference

Low-bit floating-point formats and semi-structured sparsity are increasingly supported by modern accelerators, yet combining them for LLM activation compression remains challenging: activations contain input-dependent outliers that dominate block scales in FP4 quantization, and directly applying N:M sparsity masks discards moderate values, coupling sparsification loss with quantization error. We introduce SharQ, a training-free inference method that bridges activation sparsity and FP4 quantization through an online sparse--dense decomposition. For each activation tensor, SharQ generates an input-adaptive N:M mask to extract an outlier-dominated sparse backbone, quantizes it to FP4, and defines a dense residual relative to the quantized sparse backbone rather than the unquantized sparse values. A sparse FP4 GEMM processes the backbone while a dense FP4 GEMM compensates for both mask-induced activation loss and sparse-path quantization error. The two paths share a single FP4 weight payload with path-specific scale views, and a fused preparation kernel absorbs mask generation, residual construction, and layer normalization into one operator. SharQ requires no calibration data, retraining, or model-specific tuning. Evaluated on Llama-3.1-8B, Qwen2.5-7B, Qwen3-30B-A3B, and Qwen3-VL-8B, SharQ recovers 43--63% of the NVFP4-to-FP16 accuracy gap across language and vision-language tasks, and generalizes across NVFP4, HiF4, and MXFP4 formats. On an RTX 5090, SharQ delivers 2.2--2.4

\times

latency reduction over FP16 and 1.2--1.4

\times

throughput improvement over FP8 in language model serving, and up to 1.58

\times

speedup on Wan2.2-T2V-A14B video generation when combined with SageAttention. Our code is available at https://github.com/actypedef/SharQ.

Haoqian Meng, Yilun Luo, Yafei Zhao +4

School of Computer Science and Technology, Tianjin University

4-Bit Quantization LLM Inference Acceleration Activation Quantization Neural Network Quantization

Jun 25, 2026cs.LG

What Survives When You Compress a Recursive Reasoner for the Edge?

Recursive reasoning models solve hard structured tasks with a few million parameters by iterating a latent state, but deploying them on edge hardware means compressing them -- and quantization noise compounds across recursive cycles rather than accumulating over output tokens, so single-pass intuitions fail. Here, we ask what survives. Across a full precision sweep and three tasks, aggressive compression preserves local prediction but destroys global reasoning: pruning, distillation, and linear attention drive puzzle-exact accuracy to zero while cell accuracy holds. Naive INT4 is different: its damage is architectural, striking MLP-mixing recursion but not attention on the same task, and per-channel calibration reverses it without retraining, where quantization-aware training does not. We also introduce carry-trajectory fidelity, the cosine similarity to the full-precision reasoning path, as a label-free signal that predicts both the damage and the recovery before any task evaluation, with a threshold that transfers across tasks. The result is a deployment recipe: flash-streamed embeddings remove a 99.4 MB bottleneck, one recursive cycle matches full-depth accuracy at 6x fewer FLOPs, and calibrated INT4 brings the backbone to 3.3 MB, inside a 4 MB budget.

Pearse Jim, Steven Kolawole, Opegbemi Matthias Busoye +2

ML Collective · Carnegie Mellon University

Model Compression Efficient Neural Network Inference Structured Reasoning Neural Network Quantization Neural Network Pruning

Jun 24, 2026cs.LG

Hierarchical Reinforcement Learning for Neural Network Compression (HiReLC): Pruning and Quantization

We present HiReLC, a hierarchical ensemble-reinforcement learning framework for automated joint quantization and structured pruning of deep neural networks. The framework decomposes the compression search across two levels of abstraction: low-level agents (LLAs) operate independently per block, selecting per-kernel configurations over a multi-discrete action space spanning bitwidth, pruning keep-ratio, quantization type, and granularity, while high-level agents (HLAs) coordinate global budget allocation via ensemble voting guided by Fisher Information-based sensitivity estimates. To mitigate the computational cost of policy evaluation, an iterative active learning loop interleaves surrogate-guided RL optimization with post-compression fine-tuning, using a lightweight MLP surrogate to amortize expensive evaluations and a logit-MSE proxy during cold-start. The surrogate is used for reward shaping rather than as a replacement for final post-compression evaluation. The controller is architecture-agnostic by design, with a modular layer abstraction decoupling the RL environment from the underlying network topology. Experiments across Vision Transformer and CNN benchmarks demonstrate effective parameter-storage compression ratios of 5.99 - 6.72

\times

with a 3.83 % gain in one setting and 0.55 - 5.62 % accuracy drops elsewhere, supporting hierarchical policy decomposition and sensitivity-aware guidance as practical design choices for joint neural network compression.

Kamar Hibatallah Baghdadi, Kawther Guoual Belhamidi, Sara Belhadj +2

aNational School of Artificial Intelligence (ENSIA), Sidi Abdellah Campus, Algiers, Algeria · bCosys-Grettia, Univ Gustave Eiffel, F-77454 Marne-la-Vallee, France. · aNational School of Artificial Intelligence (ENSIA), Sidi Abdellah

Hierarchical RL Neural Network Quantization Neural Network Compression Neural Network Pruning

Jun 23, 2026cs.CV

Neural Network Quantization by Learning Low-Loss Subspaces

Neural network quantization aims to find a discrete representation of parameters that preserves the performance of a full-precision (FP) model as faithfully as possible. Enforcing discrete constraints perturbs parameters away from a well-optimized minimum, generally resulting in performance degradation. Recent studies indicate that low-loss FP solutions are not isolated, but instead belong to connected low-loss subspaces of the loss landscape, where the loss maintains nearly the same minimum value. Models sampled from these subspaces are diverse and retain high accuracy. This raises the question: can a quantized model be constructed to lie within a low-loss subspace of the FP model, thereby automatically preserving performance? We address this question by learning quantization-aware linear paths in weight space optimized to minimize loss. We demonstrate that the midpoint of the resulting subspace is, by design, quantization-friendly and that its direct quantization yields performance comparable to that of quantization-aware training. The proposed procedure offers a novel perspective on weight quantization and, in contrast to conventional methods, neither relies on the straight-through estimator nor involves explicit discretization during training.

Vladimir Protsenko, Mikhalina Kharkevich, Alexander Vashchilko +1

Huawei, Montreal, Canada

Neural Network Optimization Neural Network Quantization

Jun 23, 2026cs.LG

RoPE-Aware Bit Allocation for KV-Cache Quantization

Existing low-bit KV-cache quantizers often treat each cached key as a flat vector. Under RoPE, however, a key's contribution to a future attention logit decomposes into a position-dependent sum over two-dimensional frequency blocks. This makes key-cache quantization a block-wise bit-allocation problem: high-energy RoPE blocks are more sensitive to quantization error and should receive more bits. We introduce Block-GTQ, a RoPE-aware bit allocator for key-cache quantization built on TurboQuant-MSE(TQ-MSE). For each layer and KV head, Block-GTQ computes a label-free energy score for each RoPE block and greedily allocates integer bit widths by marginal gain. Under matched K/V bit budgets, Block-GTQ better preserves RoPE query-key logits on a ten-model diagnostic panel, cutting per-layer MAE by 32-80% at 2 and 3 b/dim K-only quantization and winning all 367/367 layer comparisons against uniform TQ-MSE. These fidelity gains translate to stronger downstream long-context retrieval, understanding, and reasoning. At K2V2 on Llama-3.1-8B-Instruct, Block-GTQ raises the six-task NIAH average from 70.6 to 97.4, and the LongBench-EN average from 36.87 to 53.31. On AIME 2024/2025 with DeepSeek-R1-Distill-Qwen-7B, without an fp16 recent-key buffer, Block-GTQ at K3V2 scores 51.7/37.5, close to fp16's 54.2/37.9, whereas uniform TQ-MSE collapses to 0.0/0.0. We further implement a packed-cache serving path. On a single H800 GPU with Qwen2.5-3B-Instruct, packed K3V3 achieves 3.24x KV-cache compression with fp16-comparable quality, runs 1.34x faster than fp16 FlashAttention2 at 128K context, reduces peak memory from 56.31 GB to 19.85 GB, and remains feasible at 256K and 512K where fp16 OOMs. Code is available at https://github.com/JIA-Lab-research/blockgtq.

Fengfeng Liang, Yuechen Zhang, Jiaya Jia

1Hong Kong University of Science and Technology · 2The Chinese University of Hong Kong · 3MiMo, Xiaomi Corporation

KV Caching LLM Inference Acceleration Attention Quantization KV-Cache Quantization Low-Bit Quantization Neural Network Quantization

Jun 22, 2026cs.LG

HyperQuant: A Rate-Distortion-Optimal Quantization Pipeline for Large Language and Diffusion Models

We present HyperQuant (Hadamard, optimallY Packing, Entropy Rice-coding), a unified post-training quantization pipeline for the weights and the KV cache of large language and diffusion transformers. Across a suite of self-contained experiments (Table 1), HyperQuant outperforms the recent HIGGS scheme at every operating point from 3 to 5 bits per scalar (bps) on weights, and beats both TurboQuant and OCTOPUS on KV quantization down to 1.7 bps. Beyond the LLM setting, HyperQuant quantizes the 19B-parameter LTX-2 DiT video model with no observable per-frame artifacts. End-to-end on an H100 at 4 bps, HyperQuant compresses the linear weights ~3.9x and the KV cache ~3.79x at near-lossless quality. HyperQuant combines four known ideas into a single construction: (i) a per-tile Randomized Hadamard Transform that makes the per-coordinate distribution of weights and activations approximately Gaussian; (ii) quantization to a low-dimensional optimal lattice (E8, D4, A2, or Z); (iii) lossless bit-stripping and near-entropy-optimal variable-length Rice coding of the lattice indices; and (iv) bias-correction methods for the KV cache that keep the reconstruction unbiased under inner products, preserving attention semantics. We further integrate the pipeline with 8-bit and 4-bit Tensor-Core MMA paths (fp8-e4m3, int8, nvfp4, mxfp4), and find that int8 beats fp8 on the post-RHT lattice output. Project page: https://moonmath.ai/hyperquant/

Yuval Domb, Hadar Sackstein, Tomer Solberg

LLM Quantization Diffusion Model Quantization KV-Cache Quantization Post-Training Quantization Neural Network Quantization

Jun 20, 2026cs.CV

ScalePredictor: Instance-aware Scale Learning for Accurate Quantization of Vision Transformers

Vision Transformers have achieved remarkable success in many fields, yet their deployment on edge devices remains challenging due to their substantial computational demands. Post-Training Quantization (PTQ) offers an attractive solution by compressing models using a small calibration set with minimal training overhead. However, most existing PTQ works adopt a static quantization paradigm that is uniformly applied to all instances. Given the substantial diversity of natural images, the activation distributions vary significantly across samples, making these methods inherently suboptimal. In this paper, we propose ScalePredictor, a dynamic quantization framework for accurate and efficient quantization scale learning of ViTs. We first reveal a hidden correlation between the distribution range of shallow-layer activations and the optimal scales of deeper layers. Based on this, we develop a scale learning mechanism that integrates an efficient range extraction approach to capture robust range statistics at the shallow stage, which are then fed into a Taylor-motivated polynomial scale projection module to generate all quantization scales simultaneously. With the efficiency of polynomial approximation, ScalePredictor introduces insignificant computational overhead while avoiding costly just-in-time calibration. Extensive experiments on ImageNet demonstrate that ScalePredictor consistently outperforms prior PTQ methods, achieving a more favorable accuracy-efficiency trade-off. Code and additional results are shown in the supplementary materials.

Changjun Li, Runqing Jiang, Lian Xu +3

School of Electronics and Communication Engineering, Shenzhen Campus, Sun Yat-sen University, Shenzhen, China · Computer Science and Software Engineering, The University of Western Australia, Perth, Australia · University of Oxford, Oxford, United Kingdom

Vision Transformer Activation Quantization Post-Training Quantization Neural Network Quantization

Jun 17, 2026cs.LG

Displacement Is Not Direction: Evaluating Fidelity Metrics for Quantized LLM Deployment

Fidelity metrics, such as per-token KL divergence (KLD) against a high-precision reference, are often used in practice as low-cost proxies for benchmark quality. We test this practice on a 28-quant cohort of Qwen3.6-35B-A3B and a 41-quant cohort of Devstral-Small-2-24B, evaluated across a suite of downstream benchmarks. We find that KLD is strongly correlated with benchmark score over the full cohort (

ρ=-0.72

on Qwen and

ρ=-0.86

on Devstral, both with

p<0.001

). However, this relationship collapses to non-significance in the near-baseline silent zone (

ρ=+0.00

on Qwen and

ρ=-0.24

,

p=0.36

, on Devstral). This collapse persists across 14 measurement variants, including different KLD aggregations, perplexity formulations, top-1 agreement, calibration corpora, and context lengths. At the per-prompt level, KLD has only weak failure-prediction power on code, with failed-vs-passed geometric-mean ratios in

[1.08,1.22]

across five models on LiveCodeBench, and fails as a cross-model router, achieving only

42.3\%-49.4\%

accuracy on disagreement prompts. We trace the collapse to a structural decomposition: KLD primarily measures the volume of disagreement with the reference, with silent-zone composite

ρ=+0.94

(

p<0.001

) on Qwen and

+0.55

(

p=0.03

) on Devstral, while its relationship to the direction of those disagreements is weak and task-conditional.

Miloš Nikolić, Ali Hadi Zadeh, Enrique Torres Sanchez +1

ByteShape · University of Toronto · Vector Institute for Artificial Intelligence

LLM Quantization LLM Evaluation Neural Network Quantization

Jun 16, 2026cs.LG

Ternary Mamba: Grouped Quantization-Aware Training of W1.58A16 State Space Models

State Space Models (SSMs) such as Mamba-2 offer linear-time inference but their memory footprint limits edge deployment. Prior ternary SSM work (Slender-Mamba) trains from scratch on 150B tokens; we show a pretrained checkpoint suffices, reducing the marginal token budget by 1,000x. Using grouped quantization-aware training (QAT) with knowledge distillation from a frozen FP16 teacher, we compress Mamba-2 1.3B to 3.61x (2,687 to 744 MB) and achieve 48.1% zero-shot accuracy (7-task average) in just 102M tokens (4 GPU-hours, single H100) -- approaching Bi-Mamba's 48.4% (within +/-0.9pp CI). This QAT-from-pretrained setting reveals zero-ratio collapse, a novel instability caused by learnable quantization scales that does not arise in from-scratch training. We further show that post-hoc correction strategies effective for Transformers fail for SSMs due to error accumulation through the recurrence. These results demonstrate that ternary SSMs do not require expensive from-scratch training: QAT from pretrained checkpoints with KD is a data-efficient alternative.

Ramprasath Ganesaraja, Sahil Dilip Panse, Swathika N

EdgeVerve Systems Limited

LLM Quantization Ternary Quantization LLM Compression Quantization-Aware Training Neural Network Quantization State Space Models

Jun 15, 2026cs.LG

MODE: Modality-Decomposed Expert-Level Mixed-Precision Quantization for MoE Multimodal LLMs

Mixture-of-Experts Multimodal Large Language Models (MoE-MLLMs) offer remarkable performance but incur prohibitive GPU memory costs, making compression essential. Among PTQ methods, expert-level mixed-precision quantization has proven effective for MoE-LLMs, yet suffers notable degradation on MoE-MLLMs due to two overlooked biases in expert importance estimation. (1) At the cross-modal level, the numerical dominance of vision tokens causes expert selection frequency to be dominated by vision tokens, masking experts that are critical to the text modality; (2) at the intra-vision level, the large proportion of redundant vision tokens further skew frequency statistics, obscuring experts critical for informative visual content. To bridge gaps, we propose MODE, a modality-decomposed expert-level mixed-precision quantization framework for MoE-MLLMs that decomposes expert selection frequency by modality, filters redundant vision tokens to obtain denoised visual frequency, and further evaluates quantization sensitivity per modality as a complementary signal to frequency-based estimation. These signals are integrated into an Integer Linear Programming formulation to assign per-expert bit-widths under a given budget. Extensive experiments show that MODE is particularly well-suited for MoE-MLLMs, limiting average performance loss to within 2.9% at W3A16, with larger gains at the extreme 2-bit setting.

Yuanteng Chen, Peisong Wang, Zhilei Liu +9

Institute of Automation, Chinese Academy of Sciences · School of Artificial Intelligence, University of Chinese Academy of Sciences · 3Zhongguancun Academy

Mixed-Precision Quantization LLM Quantization VLM Quantization Multimodal Large Language Models Mixture-of-Experts Quantization Multimodal Foundation Models+3

Jun 14, 2026cs.LG

MosaicQuant: Inlier-Outlier Disaggregation for Unified 4-Bit LLM Quantization

4-bit quantization significantly reduces the memory footprint and accelerates the inference of large language models (LLMs). However, its limited bit-width representation struggles to faithfully capture both dense common values (\emph{inliers}) and rare large-magnitude values (\emph{outliers}), causing substantial accuracy degradation. Existing mixed-precision methods mitigate this by retaining outliers in high precision, but at the cost of breaking the uniformity of low-bit execution, introducing precision conversion and extra data movement that undermine practical speedup. We propose \textbf{MosaicQuant}, a unified 4-bit LLM quantization paradigm built on a novel principle of \emph{inlier--outlier disaggregation}. Rather than elevating outlier precision, MosaicQuant quantizes the full weight matrix into a dense 4-bit base component, where inliers are captured faithfully while outlier are inevitably quantized. A sparse 4-bit residual component is then introduced to compensate for these quantization errors, selectively targeting the most error-critical weight blocks where output distortion is shown to be concentrated. However, a unified representation alone is insufficient, as naïvely executing the sparse residual as a separate kernel still breaks the unified low-bit inference pipeline. To bridge this gap, we introduce \textbf{ZipperEngine}, which fuses sparse block computation into the dense 4-bit GEMM kernel via an overlapped pipeline, unifying not only the representation but also the execution into a single coherent low-bit inference pipeline. Extensive experiments on LLaMA3 and Qwen3 demonstrate that MosaicQuant preserves near-FP16 accuracy while achieving up to

1.24\times

speedup over the W16A16 baseline.

Yangjia Hu, Haodong Wang, Zicong Hong +8

HKUST · EPFL · MetaX Integrated Circuits Co., Ltd

LLM Quantization 4-Bit Quantization LLM Inference Acceleration Weight-Only Quantization Neural Network Quantization

Jun 14, 2026cs.NE

AQ4SViT: An Automated Quantization Framework with Search Gating Policy for Compressing Spiking Vision Transformers

Spiking Vision Transformers (SViTs) have emerged as alternative low-power ViT models, but their large sizes hinder their deployments on resource-constrained embedded AI systems. To address this, state-of-the-art works proposed quantization techniques to compress SViT models, but their manual, human-guided approach needs a huge design time and power/energy consumption to find the appropriate quantization setting for each given network, making this approach not scalable for quantizing multiple networks. Toward this, we propose AQ4SViT, a novel automated quantization framework for SViTs that can provide quick quantization settings with good trade-offs between accuracy and memory. To achieve this, AQ4SViT employs the following key ideas: quantization search strategy that evaluates the quantization setting candidates while considering the accuracy constraint; and search gating policy that quickly evaluates and selects promising quantization candidates by leveraging membrane potential drift as a performance proxy. In the search gating policy, AQSViT employs two search algorithm variants to provide trade-off options: Greedy search, which performs fast but may lead to local optima; and Beam search, which performs slower but has better performance in finding global optima selection due to a wider search space. Experimental results show that AQ4SViT-Greedy quickly finds the appropriate quantization settings, achieving up to 6.6x faster search time and up to 82.5% memory saving compared to the state-of-the-art; while AQ4SViT-Beam further reduces the memory footprint by up to 90% compared to the state-of-the-art, but with 4.5x longer search time; all these results are obtained while maintaining high accuracy within 1.5% from the original/non-quantized models on the ImageNet dataset. These results highlight that AQ4SViT framework offers advancements toward SViT deployments on embedded AI systems.

Rachmad Vidya Wicaksana Putra, Saad Iftikhar, Muhammad Shafique

eBRAIN Lab, Division of Engineering, New York University (NYU) Abu Dhabi, United Arab Emirates (UAE)

Spiking Transformers Neural Network Quantization Neural Network Compression

Jun 12, 2026cs.LG

Signed Symmetric Quantization for Few-Bit Integers

The signed integer alphabet contains one more negative representable value than positive. Yet, by convention, the standard symmetric integer quantizer fixes its scale to be strictly positive, which assigns this extra representable value to the negative tail and can force clipping of positive outliers. In this work, we show that, at few-bit precision, such clipping is a non-trivial source of quantization error. Asymmetric quantization addresses this problem with a zero point, shifting the grid toward the observed data range; however, this flexibility is well-known to carry a runtime penalty. For example, in llama.cpp on an AMD EPYC(TM) "Turin" CPU, a 4-bit symmetric format uses up to 9% less memory with up to 2.45

\times

higher throughput than its asymmetric counterpart. We highlight signed symmetric quantization as a third option that retains the runtime profile of symmetric quantization without the penalty of the asymmetric format: our signed absmax grid places the extra representable value on the dominant-outlier tail through a principled and lightweight sign selection rule while keeping the zero point at zero. Our theoretical analysis offers two main results. First, we establish the signed absmax grid as conditionally bound-optimal on

\ell_2

quantization error, and show that the condition holds for 88-99% of weight groups across pre-trained large language models (LLMs) at low bit widths. Second, we show that negating the scale of a standard symmetric quantizer is analytically equivalent to a unit zero point shift on the same signed integer alphabet. We empirically validate our proposal on models from the Qwen3, Qwen3.5, and Llama3 families, and observe improvement in perplexity and downstream few-shot accuracy over the standard unsigned symmetric quantizer at no extra inference cost

Ian Colbert, Eashan Dash, Pablo Monteagudo-Lago +5

AMD

LLM Quantization Weight-Only Quantization Post-Training Quantization Low-Bit Quantization Neural Network Quantization

Jun 11, 2026cs.LG

Quantizing Time-Series Models As Dynamical Systems: Trajectory-Based Quantization Sensitivity Score

We introduce the Trajectory-based Quantization Sensitivity Score (TQS), a metric that reframes post-training quantization (PTQ) through the lens of dynamical-systems stability. By modeling the network's rollout as a discrete-time dynamical system, TQS characterizes how quantization-induced errors propagate and amplify over the rollout horizon. Unlike conventional PTQ methods, where sensitivity analysis is often coupled to the quantization procedure, TQS enables a priori sensitivity estimation decoupled from quantizer selection and bit-width assignment. This separation allows for quantization budget planning even for black-box or compiled networks with fused operators. Building on this, we present TQS-PTQ, a flexible mixed-precision framework that requires no calibration data or costly second-order approximations. Our experiments show that a dynamical-systems perspective provides a robust, high-performing pathway for low-precision deployment in resource-constrained settings.

Mariya Pavlova, Harrison Bo Hua Zhu, Lidia Vitanova +2

Department of Computing, Imperial College London, London, United Kingdom · Department of Public Health, University of Copenhagen, Copenhagen, Denmark · GATE Institute, Sofia University St. Kliment Ohridski, Sofia, Bulgaria +1

Mixed-Precision Quantization Post-Training Quantization Neural Network Quantization

Jun 11, 2026cs.LG

TWLA: Achieving Ternary Weights and Low-Bit Activations for LLMs via Post-Training Quantization

Large language models (LLMs) exhibit exceptional general language processing capabilities, but their memory and compute costs hinder deployment. Ternarization has emerged as a promising compression technique, offering significant reductions in model size and inference complexity. However, existing methods struggle with heavy-tailed activation distributions and therefore keep activations in high precision, fundamentally limiting end-to-end inference acceleration. To overcome this limitation, we propose TWLA, a post-training quantization (PTQ) framework that achieves 1.58-bit weight compression and 4-bit activation quantization while maintaining high accuracy. TWLA comprises three components: (1) Euclidean-to-Manifold Asymmetric Ternary Quantizer (E2M-ATQ) minimizes layer-output error under weight ternarization via a two-stage optimization from Euclidean initialization to manifold relocation; (2) Kronecker Orthogonal Tri-Modal Shaping (KOTMS) applies a Kronecker-structured orthogonal rotation to reshape weights into ternary-friendly tri-modal distributions, while the shared rotation statistically suppresses activation outliers; and (3) Inter-Layer Aware Activation Mixed Precision (ILA-AMP) explicitly introduces adjacent-layer second-order interaction costs in bit allocation and jointly optimizes for the layer-wise disparity of activation quantization gains induced by the shared orthogonal transform, preventing cascades triggered by a few weak layers. Extensive experiments demonstrate that TWLA maintains high accuracy under W1.58A4, while delivering significant inference acceleration. The code is available at https://github.com/Kishon-zzx/TWLA.

Zhixiong Zhao, Zukang Xu, Zhixuan Chen +3

Houmo AI, China · 2Southeast University, Nanjing, China

Mixed-Precision Quantization LLM Quantization Ternary Quantization Activation Quantization Post-Training Quantization Neural Network Quantization

Jun 11, 2026cs.LG

Multi-Bitwidth Quantization for LLMs Using Additive Codebooks

As large language models (LLMs) are increasingly deployed across heterogeneous hardware with varying resource constraints, the ability to adaptively manage the trade-off between performance and efficiency without retraining is critical. We propose Drop-by-Drop, a novel multi-bitwidth post-training quantization framework that enables inference-time precision control over LLM weights from a single trained model. Our method is theoretically grounded in information theory and successive refinement. We establish that LLM weights, which commonly follow a Gaussian distribution, can be optimally reconstructed with increasing fidelity as additional bits are incorporated, under a weighted mean squared error distortion motivated by LLM loss functions. To realize this in practice, Drop-by-Drop incorporates Matryoshka-style supervision into the loss function, exploiting the structure of additive codebooks. Drop-by-Drop produces a single model where ordered subsets of codebooks yield accurate partial reconstructions at each precision level. This approach significantly reduces storage and memory overhead by allowing a single checkpoint to serve multiple bitwidths, while maintaining competitive perplexity and accuracy across major architectures, such as Qwen, LLaMA, Gemma, and Mistral.

Liza Babaoglu, Shuangyi Chen, Ashish Khisti

University of Toronto

LLM Quantization Weight-Only Quantization Iterative Refinement Post-Training Quantization Neural Network Quantization