Gaussian Weight Sampling for Scalable, Efficient and Stable Pseudo-Quantization Training

• URL: http://arxiv.org/abs/2505.11170v1
• Date: Fri, 16 May 2025 12:14:12 GMT
• Title: Gaussian Weight Sampling for Scalable, Efficient and Stable Pseudo-Quantization Training
• Authors: Myeonghwan Ahn, Sungjoo Yoo,
• Abstract summary: Pseudo-quantization training (PQT) addresses the issues of fully quantized training (FQT)<n>We show that PQT with weight sampling is (3) stable while pre-training GPT2 and Llama2 models with up to 1B parameters and 300B tokens.
• Score: 5.908471365011943
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Ever-growing scale of large language models (LLMs) is pushing for improved efficiency, favoring fully quantized training (FQT) over BF16. While FQT accelerates training, it faces consistency challenges and requires searching over an exponential number of cases, each needing over 200B tokens to ensure stability. Pseudo-quantization training (PQT) addresses the issues of FQT, although it is not well-studied. We explore the practical implications of PQT in detail and propose a noise distribution $R$ that is floating-point (FP)-friendly, with ideal properties including stochastic precision annealing. As a result, the proposed method serves as an effective theoretical foundation for low-precision FP parameters through PQT, utilizing efficient fake quantization via an addition and subsequent FP casting. We demonstrate that Gaussian weight sampling is (1) scalable: supports low-precision FP parameters down to FP6 and high-precision noise up to 9-bit with BF16 operator. The proposed method is (2) efficient: incurring computational overhead as low as 1.40\% on the A100 GPU in terms of Llama2 training tokens per second, and requiring 2 bytes per parameter in GPU memory. We demonstrate that PQT with Gaussian weight sampling is (3) stable: closely following or even surpassing performance of the BF16 baseline while pre-training GPT2 and Llama2 models with up to 1B parameters and 300B tokens.

Related papers

• PoTPTQ: A Two-step Power-of-Two Post-training for LLMs [27.141872509108122]
  Large Language Models (LLMs) have demonstrated remarkable performance across various natural language processing (NLP) tasks.<n>Power-of-two (PoT) quantization is a general tool to counteract this difficulty.<n>We propose a novel POT quantization framework for LLM weights that (i) outperforms state-of-the-art accuracy in extremely low-precision number formats, and (ii) enables faster inference through more efficient dequantization.
  arXiv  Detail & Related papers  (2025-07-16T06:44:14Z)
• FP4 All the Way: Fully Quantized Training of LLMs [26.195547788434908]
  We demonstrate fully quantized training (FQT) of large language models (LLMs) using predominantly 4-bit floating-point (FP4) precision.<n>We investigate key design choices for FP4, including block sizes, scaling formats, and rounding methods.
  arXiv  Detail & Related papers  (2025-05-25T12:14:25Z)
• PTQ1.61: Push the Real Limit of Extremely Low-Bit Post-Training Quantization Methods for Large Language Models [64.84734437930362]
  Large Language Models (LLMs) suffer severe performance degradation when facing extremely low-bit (sub 2-bit) quantization.<n>We propose an extremely low-bit PTQ method called PTQ1.61, which enables weight quantization to 1.61-bit for the first time.<n>Experiments indicate our PTQ1.61 achieves state-of-the-art performance in extremely low-bit quantization.
  arXiv  Detail & Related papers  (2025-02-18T08:04:58Z)
• HALO: Hadamard-Assisted Lower-Precision Optimization for LLMs [45.37278584462772]
  We present HALO, a novel quantization-aware training approach for Transformers.<n>Our approach ensures that all large matrix multiplications during the forward and backward passes are executed in lower precision.<n>Applying to LLAMA-family models, HALO achieves near-full-precision-equivalent results during fine-tuning on various tasks.
  arXiv  Detail & Related papers  (2025-01-05T18:41:54Z)
• EfficientQAT: Efficient Quantization-Aware Training for Large Language Models [50.525259103219256]
  quantization-aware training (QAT) offers a solution by reducing memory consumption through low-bit representations with minimal accuracy loss. We propose Efficient Quantization-Aware Training (EfficientQAT), a more feasible QAT algorithm. EfficientQAT involves two consecutive phases: Block-wise training of all parameters (Block-AP) and end-to-end training of quantization parameters (E2E-QP)
  arXiv  Detail & Related papers  (2024-07-10T17:53:30Z)
• From PEFT to DEFT: Parameter Efficient Finetuning for Reducing Activation Density in Transformers [52.199303258423306]
  We propose a novel density loss that encourages higher activation sparsity in pre-trained models. Our proposed method, textbfDEFT, can consistently reduce activation density by up to textbf44.94% on RoBERTa$_mathrmLarge$ and by textbf53.19% (encoder density) and textbf90.60% (decoder density) on Flan-T5$_mathrmXXL$.
  arXiv  Detail & Related papers  (2024-02-02T21:25:46Z)
• Trainable Fixed-Point Quantization for Deep Learning Acceleration on FPGAs [30.325651150798915]
  Quantization is a crucial technique for deploying deep learning models on resource-constrained devices, such as embedded FPGAs. We present QFX, a trainable fixed-point quantization approach that automatically learns the binary-point position during model training. QFX is implemented as a PyTorch-based library that efficiently emulates fixed-point arithmetic, supported by FPGA HLS.
  arXiv  Detail & Related papers  (2024-01-31T02:18:27Z)
• QUIK: Towards End-to-End 4-Bit Inference on Generative Large Language Models [57.04178959678024]
  We show that the majority of inference computations for large generative models can be performed with both weights and activations being cast to 4 bits. We achieve this via a hybrid quantization strategy called QUIK, which compresses most of the weights and activations to 4-bit. We provide GPU kernels matching the QUIK format with highly-efficient layer-wise runtimes, which lead to practical end-to-end throughput improvements of up to 3.4x.
  arXiv  Detail & Related papers  (2023-10-13T17:15:05Z)
• A2Q: Accumulator-Aware Quantization with Guaranteed Overflow Avoidance [49.1574468325115]
  accumulator-aware quantization (A2Q) is a novel weight quantization method designed to train quantized neural networks (QNNs) to avoid overflow during inference. A2Q introduces a unique formulation inspired by weight normalization that constrains the L1-norm of model weights according to accumulator bit width bounds. We show A2Q can train QNNs for low-precision accumulators while maintaining model accuracy competitive with a floating-point baseline.
  arXiv  Detail & Related papers  (2023-08-25T17:28:58Z)
• Quantized Neural Networks for Low-Precision Accumulation with Guaranteed Overflow Avoidance [68.8204255655161]
  We introduce a quantization-aware training algorithm that guarantees avoiding numerical overflow when reducing the precision of accumulators during inference. We evaluate our algorithm across multiple quantized models that we train for different tasks, showing that our approach can reduce the precision of accumulators while maintaining model accuracy with respect to a floating-point baseline.
  arXiv  Detail & Related papers  (2023-01-31T02:46:57Z)
• GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers [34.91478831993398]
  GPTQ is a new one-shot weight quantization method based on approximate second-order information. It can quantize GPT models with 175 billion parameters in approximately four GPU hours. Our method more than doubles the compression gains relative to previously-proposed one-shot quantization methods.
  arXiv  Detail & Related papers  (2022-10-31T13:42:40Z)

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.