{
  "abstract": "Spiking Neural Networks (SNNs) are energy-efficient due to their event-driven computation and sparse spike-based communication. However, most existing SNN models adopt a uniform time-step across all layers, which limits temporal flexibility and leads to suboptimal accuracy latency trade-offs under low-latency inference. To address this limitation, we propose Mixed-Timestep Spiking Neural Networks (MT-SNNs), a paradigm in which different layers operate with different time-steps, enabling layer-wise temporal allocation for efficient inference. To construct MT-SNNs, we develop a quantization-aware ANN-to-SNN conversion framework. Specifically, we establish a theoretical relationship between ANN activation bit-width and SNN time-steps, showing that the SNN time-step T can be approximated by the activation quantization level 2n of the source ANN. Based on this relationship, layer-wise activation bit-widths in ANNs can be mapped to layer-wise time-steps in SNNs. We then jointly optimize quantization levels and firing thresholds to obtain compact mixed-timestep configurations that reduce inference latency while preserving accuracy. Finally, we identify the temporal dimension mismatch issue caused by heterogeneous layer-wise time-steps and propose Scaled Synaptic Current Accumulation (SSCA), a fused temporal alignment mechanism that aligns inter-layer propagation in the synaptic-current domain without instantiating a standalone temporal-alignment kernel. Extensive experiments on CIFAR-10, CIFAR-100, ImageNet-1K, CIFAR10-DVS, and DVS-Gesture demonstrate the effectiveness of the proposed method. On ImageNet-1K, MT-SNNs based on ResNet-34 achieve 73.63% Top-1 accuracy with only 4.88 average time-steps, demonstrating a favorable accuracy latency trade-off for low-latency SNN conversion.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": "Original Research",
  "arxiv": null,
  "author_total": 7,
  "authors": [
    {
      "name": "Mingxin Guo"
    },
    {
      "name": "Xingting Yao"
    },
    {
      "name": "Qianpeng Li"
    },
    {
      "name": "Yaoyao Li"
    },
    {
      "name": "Gang Li"
    },
    {
      "name": "Jian Cheng"
    },
    {
      "name": "Liang Chen"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/e8qz4vbdfav69.bib",
  "cited_by": 0,
  "doi": "10.3389/fnins.2026.1783326",
  "fulltext_url": null,
  "id": "e8qz4vbdfav69",
  "issns": [
    "1662-453X"
  ],
  "landing_url": "https://www.frontiersin.org/articles/10.3389/fnins.2026.1783326",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "feed-journal",
      "biology/neuroscience"
    ]
  ],
  "openalex": null,
  "pdf_url": null,
  "pmcid": null,
  "pmid": null,
  "published": "2026-11-02",
  "publisher": "Frontiers Media SA",
  "ris": "https://research.draconitic.com/paper/e8qz4vbdfav69.ris",
  "sources": [
    {
      "first_seen": "2026-11-02T12:20:44Z",
      "native_id": "10.3389/fnins.2026.1783326",
      "source": "feed-frontiers",
      "url": "https://www.frontiersin.org/articles/10.3389/fnins.2026.1783326"
    },
    {
      "first_seen": "2026-11-02T12:23:25Z",
      "native_id": "10.3389/fnins.2026.1783326",
      "source": "crossref",
      "url": "https://www.frontiersin.org/articles/10.3389/fnins.2026.1783326/full"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "feed-journal:biology/neuroscience",
      "subfield": "neuroscience"
    }
  ],
  "title": "Mixed-timestep spiking neural networks: a temporal alignment framework for high-efficiency neuromorphic computing",
  "title_text": "Mixed-timestep spiking neural networks: a temporal alignment framework for high-efficiency neuromorphic computing",
  "type": "peer-reviewed",
  "updated_at": "2026-11-04T00:25:35Z",
  "url": "https://research.draconitic.com/paper/e8qz4vbdfav69",
  "venue": "Frontiers in Neuroscience",
  "versions": []
}