Pith. sign in

REVIEW 2 cited by

LMUFormer: Low Complexity Yet Powerful Spiking Model With Legendre Memory Units

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2402.04882 v1 pith:FLD4F3BQ submitted 2024-01-20 cs.NE cs.AIcs.LGcs.SDeess.AS

classification cs.NEcs.AIcs.LGcs.SDeess.AS
keywords performancelmuformermodelmodelswhilecomplexityprocessingsequence
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Transformer models have demonstrated high accuracy in numerous applications but have high complexity and lack sequential processing capability making them ill-suited for many streaming applications at the edge where devices are heavily resource-constrained. Thus motivated, many researchers have proposed reformulating the transformer models as RNN modules which modify the self-attention computation with explicit states. However, these approaches often incur significant performance degradation. The ultimate goal is to develop a model that has the following properties: parallel training, streaming and low-cost inference, and SOTA performance. In this paper, we propose a new direction to achieve this goal. We show how architectural modifications to a recurrent model can help push its performance toward Transformer models while retaining its sequential processing capability. Specifically, inspired by the recent success of Legendre Memory Units (LMU) in sequence learning tasks, we propose LMUFormer, which augments the LMU with convolutional patch embedding and convolutional channel mixer. Moreover, we present a spiking version of this architecture, which introduces the benefit of states within the patch embedding and channel mixer modules while simultaneously reducing the computing complexity. We evaluated our architectures on multiple sequence datasets. In comparison to SOTA transformer-based models within the ANN domain on the SCv2 dataset, our LMUFormer demonstrates comparable performance while necessitating a remarkable 53 times reduction in parameters and a substantial 65 times decrement in FLOPs. Additionally, owing to our model's proficiency in real-time data processing, we can achieve a 32.03% reduction in sequence length, all while incurring an inconsequential decline in performance. Our code is publicly available at https://github.com/zeyuliu1037/LMUFormer.git.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Learning-based Delay Compensation for Enhanced Control of Assistive Soft Robots

    cs.RO 2025-04 conditional novelty 5.0 of 10

    A learned predictor of the delayed state change reduces the soft robot arm's tracking error by up to 64% compared to a baseline sliding-mode controller.

  2. ASRC-SNN: Adaptive Skip Recurrent Connection Spiking Neural Network

    cs.NE 2025-05 conditional novelty 4.0 of 10

    Skip recurrent connections with per-layer learned spans mitigate gradient vanishing in recurrent spiking neural networks and improve long-sequence classification accuracy.

Pith tools