Self-Supervised Quantization-Aware Knowledge Distillation

• URL: http://arxiv.org/abs/2403.11106v1
• Date: Sun, 17 Mar 2024 06:20:28 GMT
• Title: Self-Supervised Quantization-Aware Knowledge Distillation
• Abstract summary: This paper proposes a novel Self-Supervised Quantization-Aware Knowledge Distillation (SQAKD) framework. SQAKD unifies the forward and backward dynamics of various quantization functions, making it flexible for incorporating various QAT works. A comprehensive evaluation shows that SQAKD substantially outperforms the state-of-the-art QAT and KD works for a variety of model architectures.
• Score: 5.4714555711042
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Quantization-aware training (QAT) and Knowledge Distillation (KD) are combined to achieve competitive performance in creating low-bit deep learning models. However, existing works applying KD to QAT require tedious hyper-parameter tuning to balance the weights of different loss terms, assume the availability of labeled training data, and require complex, computationally intensive training procedures for good performance. To address these limitations, this paper proposes a novel Self-Supervised Quantization-Aware Knowledge Distillation (SQAKD) framework. SQAKD first unifies the forward and backward dynamics of various quantization functions, making it flexible for incorporating various QAT works. Then it formulates QAT as a co-optimization problem that simultaneously minimizes the KL-Loss between the full-precision and low-bit models for KD and the discretization error for quantization, without supervision from labels. A comprehensive evaluation shows that SQAKD substantially outperforms the state-of-the-art QAT and KD works for a variety of model architectures. Our code is at: https://github.com/kaiqi123/SQAKD.git.

Related papers

• A Target-Centric Survey of Quantization-Aware Training [53.52583463060327]
  Quantization-Aware Training (QAT) techniques have emerged as a promising solution to address these challenges.<n>QAT techniques explicitly simulate quantization effects during model training, yielding low-bit models that achieve accuracy comparable to their full-precision counterparts.
  arXiv  Detail & Related papers  (2026-08-30T09:04:47Z)
• SQuaT: Self-Supervised Knowledge Distillation via Student-Aware Quantized Teacher Features [7.743324361747458]
  Quantization-Aware Training (QAT) enables the deployment of quantized models with minimal accuracy degradation.<n>Knowledge Distillation (KD) is a common approach to address this challenge.<n>We propose SQuaT, a label-free QAT framework with KD that theoretically eliminates this lower bound.
  arXiv  Detail & Related papers  (2026-08-11T09:29:30Z)
• StableQAT: Stable Quantization-Aware Training at Ultra-Low Bitwidths [49.94623294999562]
  Quantization-aware training (QAT) is essential for deploying large models under strict memory and latency constraints.<n>Common approaches based on the straight-through estimator (STE) or soft quantizers often suffer from mismatch, instability, or high computational overhead.<n>We propose StableQAT, a unified and efficient QAT framework that stabilizes training in ultra low-bit settings.
  arXiv  Detail & Related papers  (2026-01-27T08:00:57Z)
• What Makes Low-Bit Quantization-Aware Training Work for Reasoning LLMs? A Systematic Study [59.44848132298657]
  Post-training quantization (PTQ) usually comes with the cost of large accuracy drops, especially for reasoning tasks under low-bit settings.<n>In this study, we present a systematic empirical study of quantization-aware training (QAT) for reasoning models.
  arXiv  Detail & Related papers  (2026-01-21T11:22:29Z)
• Compute-Optimal Quantization-Aware Training [50.98555000360485]
  Quantization-aware training (QAT) is a leading technique for improving the accuracy of quantized neural networks.<n>Previous work has shown that decomposing training into a full-precision (FP) phase followed by a QAT phase yields superior accuracy.<n>We investigate how different QAT durations impact final performance.
  arXiv  Detail & Related papers  (2025-09-26T21:09:54Z)
• Punching Above Precision: Small Quantized Model Distillation with Learnable Regularizer [9.85847764731154]
  Game of Regularizer (GoR) is a learnable regularization method that adaptively balances task-specific (TS) and distillation losses.<n>GoR consistently outperforms state-of-the-art QAT-KD methods on low-power edge devices.<n>We also introduce QAT-EKD-GoR, an ensemble distillation framework that uses multiple heterogeneous teacher models.
  arXiv  Detail & Related papers  (2025-09-25T07:43:13Z)
• Data-Augmented Quantization-Aware Knowledge Distillation [1.8126132932201138]
  Quantization-aware training (QAT) and Knowledge Distillation (KD) are combined to achieve competitive performance in creating low-bit deep learning models.<n>The relationship between quantization-aware KD and data augmentation (DA) remains unexplored.<n>We propose a novel metric which evaluates DAs according to their capacity to maximize the Contextual Mutual Information.
  arXiv  Detail & Related papers  (2025-09-04T03:24:35Z)
• High-Fidelity Coherent-One-Way QKD Simulation Framework for 6G Networks: Bridging Theory and Reality [105.73011353120471]
  Quantum key distribution (QKD) has been emerged as a promising solution for guaranteeing information-theoretic security. Due to the considerable high-cost of QKD equipment, a lack of QKD communication system design tools is challenging. This paper introduces a QKD communication system design tool.
  arXiv  Detail & Related papers  (2025-01-21T11:03:59Z)
• Active Data Curation Effectively Distills Large-Scale Multimodal Models [66.23057263509027]
  Knowledge distillation (KD) is the de facto standard for compressing large-scale models into smaller ones. In this work we explore an alternative, yet simple approach -- active data curation as effective distillation for contrastive multimodal pretraining. Our simple online batch selection method, ACID, outperforms strong KD baselines across various model-, data- and compute-configurations.
  arXiv  Detail & Related papers  (2024-11-27T18:50:15Z)
• QSpec: Speculative Decoding with Complementary Quantization Schemes [37.007621357142725]
  Quantization has been substantially adopted to accelerate inference and reduce memory consumption of large language models. We propose a novel quantization paradigm called QSPEC, which seamlessly integrates two complementary quantization schemes for speculative decoding. QSPEC empirically boosts token generation throughput by up to 1.80x without any quality compromise.
  arXiv  Detail & Related papers  (2024-10-15T05:57:51Z)
• Boosting CLIP Adaptation for Image Quality Assessment via Meta-Prompt Learning and Gradient Regularization [55.09893295671917]
  This paper introduces a novel Gradient-Regulated Meta-Prompt IQA Framework (GRMP-IQA) The GRMP-IQA comprises two key modules: Meta-Prompt Pre-training Module and Quality-Aware Gradient Regularization. Experiments on five standard BIQA datasets demonstrate the superior performance to the state-of-the-art BIQA methods under limited data setting.
  arXiv  Detail & Related papers  (2024-09-09T07:26:21Z)
• QADYNAMICS: Training Dynamics-Driven Synthetic QA Diagnostic for Zero-Shot Commonsense Question Answering [48.25449258017601]
  State-of-the-art approaches fine-tune language models on QA pairs constructed from CommonSense Knowledge Bases. We propose QADYNAMICS, a training dynamics-driven framework for QA diagnostics and refinement.
  arXiv  Detail & Related papers  (2023-10-17T14:27:34Z)
• Poster: Self-Supervised Quantization-Aware Knowledge Distillation [6.463799944811755]
  Quantization-aware training (QAT) starts with a pre-trained full-precision model and performs quantization during retraining. Existing QAT works require supervision from the labels and they suffer from accuracy loss due to reduced precision. This paper proposes a novel Self-Supervised Quantization-Aware Knowledge Distillation framework (SQAKD)
  arXiv  Detail & Related papers  (2023-09-22T23:52:58Z)
• Unifying Synergies between Self-supervised Learning and Dynamic Computation [53.66628188936682]
  We present a novel perspective on the interplay between SSL and DC paradigms. We show that it is feasible to simultaneously learn a dense and gated sub-network from scratch in a SSL setting. The co-evolution during pre-training of both dense and gated encoder offers a good accuracy-efficiency trade-off.
  arXiv  Detail & Related papers  (2023-01-22T17:12:58Z)
• KDExplainer: A Task-oriented Attention Model for Explaining Knowledge Distillation [59.061835562314066]
  We introduce a novel task-oriented attention model, termed as KDExplainer, to shed light on the working mechanism underlying the vanilla KD. We also introduce a portable tool, dubbed as virtual attention module (VAM), that can be seamlessly integrated with various deep neural networks (DNNs) to enhance their performance under KD.
  arXiv  Detail & Related papers  (2021-05-10T08:15:26Z)
• Learning to Perturb Word Embeddings for Out-of-distribution QA [55.103586220757464]
  We propose a simple yet effective DA method based on a noise generator, which learns to perturb the word embedding of the input questions and context without changing their semantics. We validate the performance of the QA models trained with our word embedding on a single source dataset, on five different target domains. Notably, the model trained with ours outperforms the model trained with more than 240K artificially generated QA pairs.
  arXiv  Detail & Related papers  (2021-05-06T14:12:26Z)
• KDLSQ-BERT: A Quantized Bert Combining Knowledge Distillation with Learned Step Size Quantization [1.9786767260073905]
  transformer-based language models such as BERT have shown tremendous performance improvement for a range of natural language processing tasks. We propose a novel quantization method named KDLSQ-BERT that combines knowledge distillation (KD) with learned step size quantization (LSQ) for language model quantization.
  arXiv  Detail & Related papers  (2021-01-15T02:21:28Z)
• Once Quantization-Aware Training: High Performance Extremely Low-bit Architecture Search [112.05977301976613]
  We propose to combine Network Architecture Search methods with quantization to enjoy the merits of the two sides. We first propose the joint training of architecture and quantization with a shared step size to acquire a large number of quantized models. Then a bit-inheritance scheme is introduced to transfer the quantized models to the lower bit, which further reduces the time cost and improves the quantization accuracy.
  arXiv  Detail & Related papers  (2020-10-09T03:52:16Z)
• Stochastic Precision Ensemble: Self-Knowledge Distillation for Quantized Deep Neural Networks [27.533162215182422]
  quantization of deep neural networks (QDNNs) has been actively studied for deployment in edge devices. Recent studies employ the knowledge distillation (KD) method to improve the performance of quantized networks. In this study, we propose ensemble training for QDNNs (SPEQ)
  arXiv  Detail & Related papers  (2020-09-30T08:38:37Z)

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.