Scalable MatMul-free Language Modeling

• URL: http://arxiv.org/abs/2406.02528v7
• Date: Fri, 25 Jul 2025 22:38:22 GMT
• Title: Scalable MatMul-free Language Modeling
• Authors: Rui-Jie Zhu, Yu Zhang, Steven Abreu, Ethan Sifferman, Tyler Sheaves, Yiqiao Wang, Dustin Richmond, Sumit Bam Shrestha, Peng Zhou, Jason K. Eshraghian,
• Abstract summary: MatMul operations can be eliminated from large language models.<n>MatMul-free models, tested on models up to 2.7B parameters, are comparable to state-of-the-art pre-trained Transformers.
• Score: 9.048532540945086
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large Language Models (LLMs) have fundamentally altered how we approach scaling in machine learning. However, these models pose substantial computational and memory challenges, primarily due to the reliance on matrix multiplication (MatMul) within their attention and feed-forward (FFN) layers. We demonstrate that MatMul operations can be eliminated from LLMs while maintaining strong performance, even at billion-parameter scales. Our MatMul-free models, tested on models up to 2.7B parameters, are comparable to state-of-the-art pre-trained Transformers, and the performance gap narrows as model size increases. Our approach yields significant memory savings: a GPU-efficient implementation reduces memory consumption by up to 61% during training and over 10x during inference. When adapted for a multi-chip neuromorphic system, the model leverages asynchronous processing to achieve 4x higher throughput with 10x less energy than edge GPUs.

Related papers

• Pangu Ultra MoE: How to Train Your Big MoE on Ascend NPUs [111.69640966866059]
  Sparse large language models (LLMs) with Mixture of Experts (MoE) and close to a trillion parameters are dominating the realm of most capable language models.<n>In this paper, we aim to uncover a recipe to harness such scale on Ascend NPUs.<n>The key goals are better usage of the computing resources under the dynamic sparse model structures and materializing the expected performance gain on the actual hardware.
  arXiv  Detail & Related papers  (2025-05-07T15:46:36Z)
• CoLA: Compute-Efficient Pre-Training of LLMs via Low-Rank Activation [17.807249890437767]
  We introduce CoLA and its memory-efficient implementation, CoLA-M. We leverage the low-rank structure observed widely in model activations to reduce model size, boost model capacity and training efficiency. Experiments on LLaMA models with 60 million to 7 billion parameters show that CoLA reduces the computing cost by $bf 2pmbtimes$ and improves training throughput by $bf 1.86pmbtimes$ while maintaining full-rank level performance.
  arXiv  Detail & Related papers  (2025-02-16T01:05:16Z)
• Democratizing AI: Open-source Scalable LLM Training on GPU-based Supercomputers [65.35142508909892]
  We present a novel four-dimensional hybrid parallel algorithm implemented in a highly scalable, portable, open-source framework called AxoNN. We demonstrate fine-tuning of a 405-billion parameter LLM using AxoNN on Frontier.
  arXiv  Detail & Related papers  (2025-02-12T06:05:52Z)
• Neuromorphic Principles for Efficient Large Language Models on Intel Loihi 2 [5.213433310722838]
  Large language models (LLMs) deliver impressive performance but require large amounts of energy.<n>We present a MatMul-free LLM architecture adapted for Intel's neuromorphic processor, Loihi 2.<n>Our approach leverages Loihi 2's support for low-precision, event-driven computation and stateful processing.
  arXiv  Detail & Related papers  (2025-02-12T02:40:44Z)
• Memory Layers at Scale [67.00854080570979]
  This work takes memory layers beyond proof-of-concept, proving their utility at contemporary scale.<n>On downstream tasks, language models augmented with our improved memory layer outperform dense models with more than twice the budget, as well as mixture-of-expert models when matched for both compute and parameters.<n>We provide a fully parallelizable memory layer implementation, demonstrating scaling laws with up to 128B memory parameters, pretrained to 1 trillion tokens, comparing to base models with up to 8B parameters.
  arXiv  Detail & Related papers  (2024-12-12T23:56:57Z)
• APOLLO: SGD-like Memory, AdamW-level Performance [61.53444035835778]
  Large language models (LLMs) are notoriously memory-intensive during training. Various memory-efficient Scals have been proposed to reduce memory usage. They face critical challenges: (i) costly SVD operations; (ii) significant performance trade-offs compared to AdamW; and (iii) still substantial memory overhead to maintain competitive performance.
  arXiv  Detail & Related papers  (2024-12-06T18:55:34Z)
• Ultra-Sparse Memory Network [8.927205198458994]
  This work introduces UltraMem, incorporating large-scale, ultra-sparse memory layer to address these limitations.<n>Our approach significantly reduces inference latency while maintaining model performance.<n>In experiments, the largest UltraMem we train has 20 million memory slots.
  arXiv  Detail & Related papers  (2024-11-19T09:24:34Z)
• Enabling High-Sparsity Foundational Llama Models with Efficient Pretraining and Deployment [56.44025052765861]
  Large language models (LLMs) have revolutionized Natural Language Processing (NLP), but their size creates computational bottlenecks. We introduce a novel approach to create accurate, sparse foundational versions of performant LLMs. We show a total speedup on CPUs for sparse-quantized LLaMA models of up to 8.6x.
  arXiv  Detail & Related papers  (2024-05-06T16:03:32Z)
• Dense Training, Sparse Inference: Rethinking Training of Mixture-of-Experts Language Models [62.4691912312317]
  Mixture-of-Experts (MoE) language models can reduce computational costs by 2-4$times$ compared to dense models without sacrificing performance. We propose a hybrid dense training and sparse inference framework for MoE models (DS-MoE) which achieves strong computation and parameter efficiency.
  arXiv  Detail & Related papers  (2024-04-08T14:39:49Z)
• The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits [129.6765656933016]
  We introduce a 1-bit Large Language Models (LLMs) variant, namely BitNet b1.58. The 1.58-bit LLM defines a new scaling law and recipe for training new generations of LLMs. It enables a new paradigm and opens the door for designing specific hardware optimized for 1-bit LLMs.
  arXiv  Detail & Related papers  (2024-02-27T18:56:19Z)
• FinGPT-HPC: Efficient Pretraining and Finetuning Large Language Models for Financial Applications with High-Performance Computing [10.47214968497857]
  We present high-performance methods that exploit low-rank structures to pretrain and finetune large language models. Our methods achieve a speedup of 1.3X and a model compression ratio of 2.64X for pretaining without accuracy drop. For finetuning, our methods achieve an average accuracy increase of 6.3% and 24.0% in general tasks and financial tasks.
  arXiv  Detail & Related papers  (2024-02-21T05:03:17Z)
• Scaling Sparse Fine-Tuning to Large Language Models [67.59697720719672]
  Large Language Models (LLMs) are difficult to fully fine-tune due to their sheer number of parameters. We propose SpIEL, a novel sparse finetuning method which maintains an array of parameter indices and the deltas of these parameters relative to their pretrained values. We show that SpIEL is superior to popular parameter-efficient fine-tuning methods like LoRA in terms of performance and comparable in terms of run time.
  arXiv  Detail & Related papers  (2024-01-29T18:43:49Z)
• SliceGPT: Compress Large Language Models by Deleting Rows and Columns [27.004657436024853]
  We present SliceGPT, a new post-training sparsification scheme which replaces each weight matrix with a smaller (dense) matrix, reducing the embedding dimension of the network. We show that SliceGPT can remove up to 25% of the model parameters (including embeddings) for LLAMA2-70B OPT 66B and Phi-2 models while maintaining 99%, 99% and 90% zero-shot task performance.
  arXiv  Detail & Related papers  (2024-01-26T17:35:45Z)
• FlightLLM: Efficient Large Language Model Inference with a Complete Mapping Flow on FPGAs [23.381331567339526]
  Transformer-based Large Language Models (LLMs) have made a significant impact on various domains. This paper proposes FlightLLM, enabling efficient LLMs inference with a complete mapping flow on FPGAs. FlightLLM beats NVIDIA A100 GPU with 1.2$times$ higher throughput using the latest Versal VHK158 FPGA.
  arXiv  Detail & Related papers  (2024-01-08T13:00:53Z)
• Efficient LLM Inference on CPUs [8.802223672775844]
  Large language models (LLMs) have demonstrated remarkable performance and tremendous potential across a wide range of tasks. deploying these models has been challenging due to the astronomical amount of model parameters. We propose an effective approach that can make the deployment of LLMs more efficiently.
  arXiv  Detail & Related papers  (2023-11-01T13:08:50Z)
• FineQuant: Unlocking Efficiency with Fine-Grained Weight-Only Quantization for LLMs [9.072821427818557]
  Large Language Models (LLMs) have achieved state-of-the-art performance across various language tasks but pose challenges for practical deployment. We propose an efficient weight-only quantization method that reduces memory consumption and accelerates inference for LLMs. We evaluate our approach on large-scale open source models such as OPT-175B and internal MoE models, showcasing minimal accuracy loss while achieving up to 3.65 times higher throughput.
  arXiv  Detail & Related papers  (2023-08-16T23:57:41Z)
• Full Parameter Fine-tuning for Large Language Models with Limited Resources [55.794732214059806]
  Large Language Models (LLMs) have revolutionized Natural Language Processing (NLP) but demand massive GPU resources for training. We propose a new computation, LOw-Memory Optimization (LOMO), which fuses the gradient and the parameter update in one step to reduce memory usage.
  arXiv  Detail & Related papers  (2023-06-16T11:37:15Z)
• Fine-Tuning Language Models with Just Forward Passes [92.04219196752007]
  Fine-tuning language models (LMs) has yielded success on diverse downstream tasks, but as LMs grow in size, backpropagation requires a large amount of memory. We propose a memory-efficient zerothorder (MeZO) to operate in-place, thereby fine-tuning LMs with the same memory footprint as inference.
  arXiv  Detail & Related papers  (2023-05-27T02:28:10Z)
• LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale [80.86029795281922]
  We develop a procedure for Int8 matrix multiplication for feed-forward and attention projection layers in transformers. A 175B parameter 16/32-bit checkpoint can be loaded, converted to Int8, and used immediately without performance degradation.
  arXiv  Detail & Related papers  (2022-08-15T17:08:50Z)
• M6-10T: A Sharing-Delinking Paradigm for Efficient Multi-Trillion Parameter Pretraining [55.16088793437898]
  Training extreme-scale models requires enormous amounts of computes and memory footprint. We propose a simple training strategy called "Pseudo-to-Real" for high-memory-footprint-required large models.
  arXiv  Detail & Related papers  (2021-10-08T04:24:51Z)

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.