[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-82203-en":3,"doc-seo-82203-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},82203,2336464648746,"Skyler","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Event Burst Trigger An Availability Backdoor Attack on Event-Based SNN Object Detection","Event Burst Trigger (EBT) addresses a security gap in event-based spiking neural network (SNN) object detection under edge constraints. The work introduces an availability backdoor that poisons training data with crafted event-based triggers, forcing temporally burstlike activations during inference. These bursts generate many phantom object candidates, amplifying Non-Maximum Suppression (NMS) cost. Experiments on SpikeYOLO under a poison-only model keep mAP@0.5 nearly intact (drop \u003C0.099) while NMS latency rises up to 38×, and edge tests show increased baseline utilization and reduced scheduling slack. STRIP-based detection fails to reliably separate triggered inputs from benign data.","Event Burst Trigger: An Availability Backdoor Attack on Event-Based SNN Object Detection  \nJaesun Baek, Chanwook Lee, and Eun-Kyu Lee  \nDept. of Information and Telecommunication Eng., Incheon National University, Republic of Korea  \nEmail: {jsbaek, chan001h, [eklee](eklee}@inu.ac.kr)[}](eklee}@inu.ac.kr)[@inu.ac.kr](eklee}@inu.ac.kr)  \narXiv :2607 .09115v1 [ cs .CV] 10 Jul 2026  \nAbstract—Event-based vision and spiking neural networks (SNNs) are increasingly adopted for edge intelligence under strict latency and energy constraints. However, the vulnerability of event-based SNN object detection models to availability backdoor attacks remains insufficiently studied. This paper presents Event Burst Trigger (EBT), an availability backdoor attack targeting SNN-based object detection models. EBT injects carefully crafted event-based triggers into the training data, which induce temporally concentrated event streams during inference. These burstlike activations increase the number of phantom (i.e., spurious) object candidates, and consequently inflate the computational cost of the post-processing stage, particularly Non-Maximum Suppression (NMS). We evaluate EBT on SpikeYOLO, the stateof-the-art SNN-based object detector, under a poison-only threat model that does not require modifications to the model architecture, loss function, or inference pipeline. Experimental results show that while detection accuracy remains largely preserved, with mAP@0.5 decreasing by less than 0.099, the latency of the NMS stage increases by up to 38×. This indicates that NMS can become a dominant availability bottleneck in event-based SNN object detection. Experiments on an edge platform further show that the proposed attack elevates baseline resource utilization and reduces scheduling slack without inducing conspicuous peaks in resource usage. In addition, STRIP-based backdoor detection fails to reliably distinguish the proposed attack from benign inputs. These results characterize a previously underexplored availability backdoor threat in event-based SNN object detection systems.  \nIndex Terms—Availability Backdoor Attack, Spiking Neural Networks, Event-Based Vision, Non-Maximum Suppression  \nI. INTRODUCTION  \nReal-time vision systems operate under strict timing constraints, particularly in safety-critical domains such as autonomous driving, robotics, and aerial systems [1]–[4] . To meet these constraints, inference is commonly executed on edge or onboard devices that are physically close to the sensing source, avoiding the communication overhead associated with cloud-based processing [5], [6] . While edge-based inference reduces communication latency, it is subject to limited computational and memory resources. Consequently, real-time vision systems deployed on edge platforms must account not only for detection accuracy, but also for predictable execution behavior under constrained resource [7], [8] .  \nEvent-based processing has emerged as an effective approach for addressing the resource constraints of edge platforms, and Spiking Neural Networks (SNNs), due to their  \nThis research was supported by the MSIT, Korea, under the ITRC support program (IITP-2026-RS-2023-00259061) supervised by the IITP. Corresponding author: [eklee@inu.ac.kr](eklee@inu.ac.kr).  \nasynchronous computation paradigm, are well suited for processing such data in an energy-efficient manner [9], [10] . By aligning computation with sparse event streams, SNN-based models enable low-power inference suitable for resourceconstrained environments [11] . SpikeYOLO is a representative SNN-based object detection model that achieves competitive detection performance while reducing energy consumption [12] . This combination of event-based sensing and SNNbased inference supports real-time object detection on edge devices without relying on large-scale cloud resources.  \nDespite these advantages, SNN-based object detection pipelines introduce security considerations that have received limited","cbCaisHoBgGyTVRn","https://ap.wps.com/l/cbCaisHoBgGyTVRn","pdf",500868,4,1,7,"English","en",105,"# Abstract\n# Introduction\n## Edge inference constraints\n## Event-based SNN object detection and security considerations\n## Availability backdoor attacks and NMS amplification\n## Proposed Event Burst Trigger (EBT) and contributions","[{\"question\":\"What is the Event Burst Trigger (EBT) attack?\",\"answer\":\"EBT is an availability backdoor attack that injects crafted event-based triggers into training data so that, during inference, temporally concentrated event bursts occur and increase computational load.\"},{\"question\":\"How does EBT affect object detection results and latency?\",\"answer\":\"Detection accuracy is largely preserved, with mAP@0.5 decreasing by less than 0.099, while the NMS stage latency increases by up to 38× due to the rise in phantom object candidates.\"},{\"question\":\"Why does STRIP fail to detect the proposed backdoor reliably?\",\"answer\":\"The experiments indicate that STRIP-based backdoor detection cannot reliably distinguish the event-based triggers from benign inputs, so accuracy-centric or entropybased checks do not expose the availability vulnerability.\"}]",1784178791,18,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"event-burst-trigger-an-availability-backdoor-attack-on-event-based-snn-object-detection","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":20},"https://docshare.wps.com/document/event-burst-trigger-an-availability-backdoor-attack-on-event-based-snn-object-detection/82203/",{"url":52,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-21","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What is the Event Burst Trigger (EBT) attack?","Question",{"text":75,"@type":76},"EBT is an availability backdoor attack that injects crafted event-based triggers into training data so that, during inference, temporally concentrated event bursts occur and increase computational load.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does EBT affect object detection results and latency?",{"text":80,"@type":76},"Detection accuracy is largely preserved, with mAP@0.5 decreasing by less than 0.099, while the NMS stage latency increases by up to 38× due to the rise in phantom object candidates.",{"name":82,"@type":73,"acceptedAnswer":83},"Why does STRIP fail to detect the proposed backdoor reliably?",{"text":84,"@type":76},"The experiments indicate that STRIP-based backdoor detection cannot reliably distinguish the event-based triggers from benign inputs, so accuracy-centric or entropybased checks do not expose the availability vulnerability.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,119,122,127,130,134],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":22,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]