Neural Weight Compression for Language Models

• URL: http://arxiv.org/abs/2510.11234v1
• Date: Mon, 13 Oct 2025 10:16:20 GMT
• Title: Neural Weight Compression for Language Models
• Authors: Jegwang Ryu, Minkyu Kim, Seungjun Shin, Hee Min Choi, Dokwan Oh, Jaeho Lee,
• Abstract summary: We introduce Neural Weight Compression (NWC), a novel autoencoder-based neural model weight compression algorithm.<n>NWC achieves competitive or state-of-the-art accuracy-compression tradeoffs, with particularly strong results at 4-6 bit precisions.
• Score: 17.47037037010811
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The efficient storage and transmission of language model weights is becoming increasingly important, as their scale and adoption continue to grow. However, as our understanding of this new data modality is limited, designing a good compression algorithm for language model weights heavily relies on manual, trial-and-error approaches. In this paper, we propose a learned compression framework that trains neural codecs directly from pretrained language model weights. Unlike conventional data (e.g., images), language model weights pose unique challenges: the sizes and shapes of weight tensors vary significantly, and the reconstruction quality must be judged by downstream model predictions rather than na\"ive MSE loss. To address this, we introduce Neural Weight Compression (NWC), a novel autoencoder-based neural codec tailored to model weight compression. The proposed method inherits the advantages of autoencoder-based codecs while incorporating three technical components: (1) column-wise tensor chunking and normalization; (2) an importance-aware training loss; (3) an inference-time error compensation mechanism guided by model outputs. Experiments on open-weight language models show that NWC achieves competitive or state-of-the-art accuracy-compression tradeoffs, with particularly strong results at 4-6 bit precisions where accuracy remains nearly on par with FP16 models.

Related papers

• Efficient Training with Denoised Neural Weights [65.14892033932895]
  This work takes a novel step towards building a weight generator to synthesize the neural weights for initialization. We use the image-to-image translation task with generative adversarial networks (GANs) as an example due to the ease of collecting model weights. By initializing the image translation model with the denoised weights predicted by our diffusion model, the training requires only 43.3 seconds.
  arXiv  Detail & Related papers  (2024-07-16T17:59:42Z)
• CompactifAI: Extreme Compression of Large Language Models using Quantum-Inspired Tensor Networks [1.5199992713356987]
  This paper introduces CompactifAI, an innovative compression approach using quantum-inspired networks. Our method is versatile and can be implemented with - or on top of - other compression techniques. As a benchmark, we demonstrate that a combination of CompactifAI with quantization allows to reduce a 93% memory size of LlaMA 7B.
  arXiv  Detail & Related papers  (2024-01-25T11:45:21Z)
• Activations and Gradients Compression for Model-Parallel Training [85.99744701008802]
  We study how simultaneous compression of activations and gradients in model-parallel distributed training setup affects convergence. We find that gradients require milder compression rates than activations. Experiments also show that models trained with TopK perform well only when compression is also applied during inference.
  arXiv  Detail & Related papers  (2024-01-15T15:54:54Z)
• Deep learning model compression using network sensitivity and gradients [3.52359746858894]
  We present model compression algorithms for both non-retraining and retraining conditions. In the first case, we propose the Bin & Quant algorithm for compression of the deep learning models using the sensitivity of the network parameters. In the second case, we propose our novel gradient-weighted k-means clustering algorithm (GWK)
  arXiv  Detail & Related papers  (2022-10-11T03:02:40Z)
• CrAM: A Compression-Aware Minimizer [103.29159003723815]
  We propose a new compression-aware minimizer dubbed CrAM that modifies the optimization step in a principled way. CrAM produces dense models that can be more accurate than the standard SGD/Adam-based baselines, but which are stable under weight pruning. CrAM can produce sparse models which perform well for transfer learning, and it also works for semi-structured 2:4 pruning patterns supported by GPU hardware.
  arXiv  Detail & Related papers  (2022-07-28T16:13:28Z)
• What Do Compressed Multilingual Machine Translation Models Forget? [102.50127671423752]
  We show that the performance of under-represented languages drops significantly, while the average BLEU metric only slightly decreases. We demonstrate that compression amplifies intrinsic gender and semantic biases, even in high-resource languages.
  arXiv  Detail & Related papers  (2022-05-22T13:54:44Z)
• Online Model Compression for Federated Learning with Large Models [8.48327410170884]
  Online Model Compression (OMC) is a framework that stores model parameters in a compressed format and decompresses them only when needed. OMC can reduce memory usage and communication cost of model parameters by up to 59% while attaining comparable accuracy and training speed when compared with full-precision training.
  arXiv  Detail & Related papers  (2022-05-06T22:43:03Z)
• Load-balanced Gather-scatter Patterns for Sparse Deep Neural Networks [20.374784902476318]
  Pruning, as a method to introduce zeros to model weights, has shown to be an effective method to provide good trade-offs between model accuracy and computation efficiency. Some modern processors are equipped with fast on-chip scratchpad memories and gather/scatter engines that perform indirect load and store operations on such memories. In this work, we propose a set of novel sparse patterns, named gather-scatter (GS) patterns, to utilize the scratchpad memories and gather/scatter engines to speed up neural network inferences.
  arXiv  Detail & Related papers  (2021-12-20T22:55:45Z)
• Prune Once for All: Sparse Pre-Trained Language Models [0.6063525456640462]
  We present a new method for training sparse pre-trained Transformer language models by integrating weight pruning and model distillation. These sparse pre-trained models can be used to transfer learning for a wide range of tasks while maintaining their sparsity pattern. We show how the compressed sparse pre-trained models we trained transfer their knowledge to five different downstream natural language tasks with minimal accuracy loss.
  arXiv  Detail & Related papers  (2021-11-10T15:52:40Z)
• LCS: Learning Compressible Subspaces for Adaptive Network Compression at Inference Time [57.52251547365967]
  We propose a method for training a "compressible subspace" of neural networks that contains a fine-grained spectrum of models. We present results for achieving arbitrarily fine-grained accuracy-efficiency trade-offs at inference time for structured and unstructured sparsity. Our algorithm extends to quantization at variable bit widths, achieving accuracy on par with individually trained networks.
  arXiv  Detail & Related papers  (2021-10-08T17:03:34Z)
• Training with Quantization Noise for Extreme Model Compression [57.51832088938618]
  We tackle the problem of producing compact models, maximizing their accuracy for a given model size. A standard solution is to train networks with Quantization Aware Training, where the weights are quantized during training and the gradients approximated with the Straight-Through Estimator. In this paper, we extend this approach to work beyond int8 fixed-point quantization with extreme compression methods.
  arXiv  Detail & Related papers  (2020-04-15T20:10:53Z)

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.