{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/qkformer-hierarchical-spiking-transformer","title":"QKFormer: Hierarchical Spiking Transformer using Q-K Attention","arxiv_id":"2403.16552","date":"2024-03-25","proceeding":null,"authors":["Chenlin Zhou","Han Zhang","Zhaokun Zhou","Liutao Yu","Liwei Huang","Xiaopeng Fan","Li Yuan","Zhengyu Ma","Huihui Zhou","Yonghong Tian"],"abstract":"Spiking Transformers, which integrate Spiking Neural Networks (SNNs) with Transformer architectures, have attracted significant attention due to their potential for energy efficiency and high performance. However, existing models in this domain still suffer from suboptimal performance. We introduce several innovations to improve the performance: i) We propose a novel spike-form Q-K attention mechanism, tailored for SNNs, which efficiently models the importance of token or channel dimensions through binary vectors with linear complexity. ii) We incorporate the hierarchical structure, which significantly benefits the performance of both the brain and artificial neural networks, into spiking transformers to obtain multi-scale spiking representation. iii) We design a versatile and powerful patch embedding module with a deformed shortcut specifically for spiking transformers. Together, we develop QKFormer, a hierarchical spiking transformer based on Q-K attention with direct training. QKFormer shows significantly superior performance over existing state-of-the-art SNN models on various mainstream datasets. Notably, with comparable size to Spikformer (66.34 M, 74.81%), QKFormer (64.96 M) achieves a groundbreaking top-1 accuracy of 85.65% on ImageNet-1k, substantially outperforming Spikformer by 10.84%. To our best knowledge, this is the first time that directly training SNNs have exceeded 85% accuracy on ImageNet-1K. The code and models are publicly available at https://github.com/zhouchenlin2096/QKFormer","url_abs":"https://arxiv.org/abs/2403.16552v2","url_pdf":"https://arxiv.org/pdf/2403.16552v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"qkformer-hierarchical-spiking-transformer","repo_url":"https://github.com/zhouchenlin2096/qkformer","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"qkformer-hierarchical-spiking-transformer","repo_url":"https://github.com/Fancyssc/Spiking-Transformers","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"snn","method_name":"SNN"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2403.16552","atlas_url":"https://app.syntology.ai/?focus=2403.16552","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2403.16552"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"deterministic:regex_extraction","url":"https://github.com/zhouchenlin2096/QKFormer","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/zhouchenlin2096/qkformer","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Fancyssc/Spiking-Transformers","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_fixture":1,"ran":2,"unverified":2},"by_repo_kind":{"official":{"samples":2,"ran":2,"repositories":1},"listed":{"samples":3,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":2,"samples":[{"code_sha256_prefix":"4134b8420b4c48b5","entry":"accuracy","repo":"zhouchenlin2096/QKFormer","repo_kind":"official","path":"cifar10-dvs/utils.py","file_url":"https://github.com/zhouchenlin2096/QKFormer/blob/HEAD/cifar10-dvs/utils.py","link_basis":"plan_row","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"4134b8420b4c48b5"}},{"code_sha256_prefix":"aeded02f9fc26708","entry":"build_transform","repo":"Fancyssc/Spiking-Transformers","repo_kind":"listed","path":"datasets.py","file_url":"https://github.com/Fancyssc/Spiking-Transformers/blob/HEAD/datasets.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"aeded02f9fc26708"}},{"code_sha256_prefix":"68edbf1d7855d1bc","entry":"unpack_len1_tuple","repo":"zhouchenlin2096/QKFormer","repo_kind":"official","path":"cifar10-dvs/monitor.py","file_url":"https://github.com/zhouchenlin2096/QKFormer/blob/HEAD/cifar10-dvs/monitor.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"68edbf1d7855d1bc"}},{"code_sha256_prefix":"99c6a11c5039db76","entry":"build_dataset","repo":"Fancyssc/Spiking-Transformers","repo_kind":"listed","path":"datasets.py","file_url":"https://github.com/Fancyssc/Spiking-Transformers/blob/HEAD/datasets.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"99c6a11c5039db76"}},{"code_sha256_prefix":"137a4ac791af3cd7","entry":"unpack_mix_param","repo":"Fancyssc/Spiking-Transformers","repo_kind":"listed","path":"datasets.py","file_url":"https://github.com/Fancyssc/Spiking-Transformers/blob/HEAD/datasets.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"137a4ac791af3cd7"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}