{"id":"f15db2cf472f0b25310408ba33c84fc679190b09ae9e0fe4fa97257cd4fc25a0","title":"Moshruba and Parsa argue spiking neural networks are harder to steal","argument":"Ayana Moshruba and Maryam Parsa report a benchmark study comparing model stealing attacks against spiking neural networks (SNNs) and conventional artificial neural networks (ANNs). Model stealing attacks use a model’s responses to queries to train a surrogate model that imitates its decisions, without access to its parameters or training data.\n\nThe authors tested four attacks on MNIST, Fashion-MNIST, and CIFAR-10. They report that, across every attack and dataset, SNN surrogates had 10% to 20% lower agreement with the target and 15% to 25% lower adversarial transferability than ANN surrogates. They attribute the difference to SNNs’ sparse, discrete spikes and time-dependent dynamics, which they say make their decision boundaries harder for ANN-based surrogates to approximate, even with large query budgets.\n\nThe authors argue that this resistance extends SNNs’ known resilience beyond attacks on training data to attempts to replicate model functionality, strengthening the case for neuromorphic architectures in security-sensitive applications. The abstract describes benchmark results, not how well this resistance holds in deployed systems. Would the reported gap persist against other model stealing attacks or on tasks beyond the three datasets tested?","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The evidence supplied is the arXiv abstract; it does not provide details of the experimental setup beyond the named attacks and datasets. The reported findings concern four attacks and three datasets; the abstract does not establish whether the results generalize to other attacks, tasks, or deployed systems. The authors attribute the resistance to SNN computation, but the supplied abstract does not provide further evidence or analysis of that explanation.","speaker":{"name":"Moshruba, Ayana, Parsa, Maryam","roleAtSource":"Article author"},"source":{"title":"Spiking Neural Networks Are Harder to Steal: A Benchmark Study of Model Stealing Attacks","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2610.36298","publishedAt":"2026-10-26","dateBasis":"Date basis not provided."},"evidence":[{"text":"Moshruba and Parsa define model stealing attacks as using a target model’s responses to train a surrogate that replicates its decisions, without access to its parameters or training data.","excerpt":"training a surrogate model that replicates the decision behavior of a target through black-box queries alone, without access to its parameters or training data","url":"https://arxiv.org/abs/2610.36298","relationship":"Source-stated"},{"text":"The authors benchmarked four representative attacks against SNNs and their ANN counterparts on MNIST, Fashion-MNIST, and CIFAR-10.","excerpt":"benchmarking four representative MSAs, namely the Papernot, Tramèr, T-RND, and color perturbation attacks, against SNNs and their ANN counterparts on MNIST, Fashion-MNIST, and CIFAR-10","url":"https://arxiv.org/abs/2610.36298","relationship":"Source-stated"},{"text":"The authors report that SNNs were consistently harder to steal across every tested attack and dataset; SNN surrogates had lower agreement with target models and lower adversarial transferability than ANN surrogates.","excerpt":"Across every attack and dataset, SNNs prove consistently more difficult to steal. Surrogates of SNNs achieve 10% to 20% lower agreement with the target model and 15% to 25% lower adversarial transferability than surrogates of ANNs.","url":"https://arxiv.org/abs/2610.36298","relationship":"Source-stated"},{"text":"The authors attribute the reported resistance to SNNs’ spike-based, time-dependent dynamics and argue that the findings strengthen the case for neuromorphic architectures in security-sensitive applications.","excerpt":"These findings extend the known resilience of SNNs beyond attacks on training data to the replication of model functionality, strengthening the case for neuromorphic architectures in security-sensitive applications.","url":"https://arxiv.org/abs/2610.36298","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2610.36298","checkedAt":"2026-11-07T08:50:08.274Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"Would the reported gap persist against other model stealing attacks or on tasks beyond the three datasets tested?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"Moshruba and Parsa argue spiking neural networks are harder to steal","summary":"Ayana Moshruba and Maryam Parsa report a benchmark study comparing model stealing attacks against spiking neural networks (SNNs) and conventional artificial neural networks (ANNs). Model stealing attacks use a model’s responses to queries to train a surrogate model that imitates its decisions, without access to its parameters or training data.\n\nThe authors tested four attacks on MNIST, Fashion-MNIST, and CIFAR-10. They report that, across every attack and dataset, SNN surrogates had 10% to 20% lower agreement with the target and 15% to 25% lower adversarial transferability than ANN surrogates. They attribute the difference to SNNs’ sparse, discrete spikes and time-dependent dynamics, which they say make their decision boundaries harder for ANN-based surrogates to approximate, even with large query budgets.\n\nThe authors argue that this resistance extends SNNs’ known resilience beyond attacks on training data to attempts to replicate model functionality, strengthening the case for neuromorphic architectures in security-sensitive applications. The abstract describes benchmark results, not how well this resistance holds in deployed systems. Would the reported gap persist against other model stealing attacks or on tasks beyond the three datasets tested?","question":"Would the reported gap persist against other model stealing attacks or on tasks beyond the three datasets tested?","claims":[{"text":"Moshruba and Parsa define model stealing attacks as using a target model’s responses to train a surrogate that replicates its decisions, without access to its parameters or training data.","excerpt":"training a surrogate model that replicates the decision behavior of a target through black-box queries alone, without access to its parameters or training data","support":"source-stated"},{"text":"The authors benchmarked four representative attacks against SNNs and their ANN counterparts on MNIST, Fashion-MNIST, and CIFAR-10.","excerpt":"benchmarking four representative MSAs, namely the Papernot, Tramèr, T-RND, and color perturbation attacks, against SNNs and their ANN counterparts on MNIST, Fashion-MNIST, and CIFAR-10","support":"source-stated"},{"text":"The authors report that SNNs were consistently harder to steal across every tested attack and dataset; SNN surrogates had lower agreement with target models and lower adversarial transferability than ANN surrogates.","excerpt":"Across every attack and dataset, SNNs prove consistently more difficult to steal. Surrogates of SNNs achieve 10% to 20% lower agreement with the target model and 15% to 25% lower adversarial transferability than surrogates of ANNs.","support":"source-stated"},{"text":"The authors attribute the reported resistance to SNNs’ spike-based, time-dependent dynamics and argue that the findings strengthen the case for neuromorphic architectures in security-sensitive applications.","excerpt":"These findings extend the known resilience of SNNs beyond attacks on training data to the replication of model functionality, strengthening the case for neuromorphic architectures in security-sensitive applications.","support":"source-stated"}],"limitations":["The evidence supplied is the arXiv abstract; it does not provide details of the experimental setup beyond the named attacks and datasets.","The reported findings concern four attacks and three datasets; the abstract does not establish whether the results generalize to other attacks, tasks, or deployed systems.","The authors attribute the resistance to SNN computation, but the supplied abstract does not provide further evidence or analysis of that explanation."],"mediaId":"0ed6edd9404f35cfd939b968b2e1975a24c3d37297797bb757c808dd261574b7","correctionReason":null,"id":"f15db2cf472f0b25310408ba33c84fc679190b09ae9e0fe4fa97257cd4fc25a0","publicationId":"1b35daa8-a32f-48cf-82ec-60c86fb883c2","version":1,"sourceUrl":"https://arxiv.org/abs/2610.36298","sourceTitle":"Spiking Neural Networks Are Harder to Steal: A Benchmark Study of Model Stealing Attacks","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-10-26","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-10-28T07:20:36.406Z","createdAt":"2026-11-07T07:47:42.813Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"0ed6edd9404f35cfd939b968b2e1975a24c3d37297797bb757c808dd261574b7","kind":"article","url":"https://arxiv.org/abs/2610.36298","sourceUrl":"https://arxiv.org/abs/2610.36298","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-10-30T23:03:35.276Z","method":"acquired_source","contentType":"text/html"},"recommendationReason":"Recommended by Jev for understanding this source; checked by Luna."}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2610.36298","checkedAt":"2026-11-07T08:50:08.274Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v5","mediaId":"0ed6edd9404f35cfd939b968b2e1975a24c3d37297797bb757c808dd261574b7","confidence":0.84,"provider":"typesafe","model":"jev-1.13.0","inputHash":"ea79bef61a080c0287518380779fe07cb14df967863aaa20fb255e3757e48f83","decidedAt":"2026-10-30T23:03:37.223Z"},"provenance":{"textHash":"c7a95ba7227ab369886d6c80ed0b0bcec57c8d5303355b7b28bf701781be9f48","observationId":"8090a367531858ec220485f91926b24e262c499772a8093cd000142add35bc14","retrievedAt":"2026-10-30T23:03:34.971Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"b5725063-5600-4373-81a5-b657cb2cbf57"},"reviewer":{"model":"gpt-6-luna","runId":"ffb20c35-728a-458e-89c6-de0e2f73e887"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-07T07:47:42.813Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Moshruba, Ayana","url":null,"kind":"unknown"},{"name":"Parsa, Maryam","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Moshruba, Ayana","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Moshruba, Ayana","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.36298","locator":"head/meta[name=author|citation_author]"},"assertionId":"297b3dac5c900fe7a2b0a91e19b0a0075b9637d88a2a44bc34ebf61f5a1c14c7:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-07T08:50:08.274Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"297b3dac5c900fe7a2b0a91e19b0a0075b9637d88a2a44bc34ebf61f5a1c14c7","sourceUrl":"https://arxiv.org/abs/2610.36298","contentSha256":"941b9a0193fc398415c104ae9ce9af057b6658a1a327d5cd448ab771de8cf7ad","textHash":"c7a95ba7227ab369886d6c80ed0b0bcec57c8d5303355b7b28bf701781be9f48","observedAt":"2026-11-07T08:50:08.274Z"}},{"actor":{"name":"Parsa, Maryam","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Parsa, Maryam","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.36298","locator":"head/meta[name=author|citation_author]"},"assertionId":"297b3dac5c900fe7a2b0a91e19b0a0075b9637d88a2a44bc34ebf61f5a1c14c7:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-07T08:50:08.274Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"297b3dac5c900fe7a2b0a91e19b0a0075b9637d88a2a44bc34ebf61f5a1c14c7","sourceUrl":"https://arxiv.org/abs/2610.36298","contentSha256":"941b9a0193fc398415c104ae9ce9af057b6658a1a327d5cd448ab771de8cf7ad","textHash":"c7a95ba7227ab369886d6c80ed0b0bcec57c8d5303355b7b28bf701781be9f48","observedAt":"2026-11-07T08:50:08.274Z"}}],"participationCoverage":{"total":2,"returned":2,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2610.36298","snapshotId":"297b3dac5c900fe7a2b0a91e19b0a0075b9637d88a2a44bc34ebf61f5a1c14c7","contentSha256":"941b9a0193fc398415c104ae9ce9af057b6658a1a327d5cd448ab771de8cf7ad","textHash":"c7a95ba7227ab369886d6c80ed0b0bcec57c8d5303355b7b28bf701781be9f48","observedAt":"2026-11-07T08:50:08.274Z","basis":"publisher-page-metadata","sourceLinks":40,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/f15db2cf472f0b25310408ba33c84fc679190b09ae9e0fe4fa97257cd4fc25a0/source"}},"canonicalUrl":"https://escapemeridian.com/readings/moshruba-and-parsa-argue-spiking-neural-networks-are-harder-to-f15db2cf","formats":{"html":"/readings/moshruba-and-parsa-argue-spiking-neural-networks-are-harder-to-f15db2cf","markdown":"/readings/moshruba-and-parsa-argue-spiking-neural-networks-are-harder-to-f15db2cf.md","json":"/readings/moshruba-and-parsa-argue-spiking-neural-networks-are-harder-to-f15db2cf.json"},"attribution":"Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.","maintainedGuideLinks":[],"readerContext":{"version":"reader-context-v1","attribution":"Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.","terms":[{"id":"term-neural-network","name":"Neural network","explanation":"A system of connected simple calculations, loosely inspired by the brain, that learns patterns from data.","referenceUrl":"https://developers.google.com/machine-learning/glossary#neural-network"},{"id":"term-benchmark","name":"Benchmark","explanation":"A defined set of tests used to compare systems. Results depend on the tasks and measures selected.","referenceUrl":"https://arxiv.org/abs/2211.09110"}],"entities":[{"id":"arxiv","name":"arXiv","kind":"organization","context":"arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.","referenceUrls":["https://arxiv.org/","https://info.arxiv.org/help/api/index.html"],"path":"/entities/arxiv","json":"/entities/arxiv.json","markdown":"/entities/arxiv.md"},{"id":"time","name":"TIME","kind":"organization","context":"TIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services.","referenceUrls":["https://time.com/"],"path":"/entities/time","json":"/entities/time.json","markdown":"/entities/time.md"}]},"researchQuestions":[],"withdrawn":false,"references":{"schemaVersion":1,"target":{"type":"reading","id":"f15db2cf472f0b25310408ba33c84fc679190b09ae9e0fe4fa97257cd4fc25a0","revision":"1b35daa8-a32f-48cf-82ec-60c86fb883c2","revisionKind":"publication-id","title":"Moshruba and Parsa argue spiking neural networks are harder to steal","url":"https://escapemeridian.com/readings/moshruba-and-parsa-argue-spiking-neural-networks-are-harder-to-f15db2cf"},"source":{"type":"source","id":"https://arxiv.org/abs/2610.36298","revision":"c7a95ba7227ab369886d6c80ed0b0bcec57c8d5303355b7b28bf701781be9f48","revisionKind":"sha256-source-text","title":"Spiking Neural Networks Are Harder to Steal: A Benchmark Study of Model Stealing Attacks","url":"https://arxiv.org/abs/2610.36298"},"status":"available","items":[],"next":null,"scope":"Explicit citations in public readings and public file attachments to this reading. Shared topics are not citations. Other artifact types need an explicit reference adapter."},"revisions":[{"id":"f15db2cf472f0b25310408ba33c84fc679190b09ae9e0fe4fa97257cd4fc25a0","title":"Moshruba and Parsa argue spiking neural networks are harder to steal","argument":"Ayana Moshruba and Maryam Parsa report a benchmark study comparing model stealing attacks against spiking neural networks (SNNs) and conventional artificial neural networks (ANNs). Model stealing attacks use a model’s responses to queries to train a surrogate model that imitates its decisions, without access to its parameters or training data.\n\nThe authors tested four attacks on MNIST, Fashion-MNIST, and CIFAR-10. They report that, across every attack and dataset, SNN surrogates had 10% to 20% lower agreement with the target and 15% to 25% lower adversarial transferability than ANN surrogates. They attribute the difference to SNNs’ sparse, discrete spikes and time-dependent dynamics, which they say make their decision boundaries harder for ANN-based surrogates to approximate, even with large query budgets.\n\nThe authors argue that this resistance extends SNNs’ known resilience beyond attacks on training data to attempts to replicate model functionality, strengthening the case for neuromorphic architectures in security-sensitive applications. The abstract describes benchmark results, not how well this resistance holds in deployed systems. Would the reported gap persist against other model stealing attacks or on tasks beyond the three datasets tested?","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The evidence supplied is the arXiv abstract; it does not provide details of the experimental setup beyond the named attacks and datasets. The reported findings concern four attacks and three datasets; the abstract does not establish whether the results generalize to other attacks, tasks, or deployed systems. The authors attribute the resistance to SNN computation, but the supplied abstract does not provide further evidence or analysis of that explanation.","speaker":{"name":"Moshruba, Ayana, Parsa, Maryam","roleAtSource":"Article author"},"source":{"title":"Spiking Neural Networks Are Harder to Steal: A Benchmark Study of Model Stealing Attacks","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2610.36298","publishedAt":"2026-10-26","dateBasis":"Date basis not provided."},"evidence":[{"text":"Moshruba and Parsa define model stealing attacks as using a target model’s responses to train a surrogate that replicates its decisions, without access to its parameters or training data.","excerpt":"training a surrogate model that replicates the decision behavior of a target through black-box queries alone, without access to its parameters or training data","url":"https://arxiv.org/abs/2610.36298","relationship":"Source-stated"},{"text":"The authors benchmarked four representative attacks against SNNs and their ANN counterparts on MNIST, Fashion-MNIST, and CIFAR-10.","excerpt":"benchmarking four representative MSAs, namely the Papernot, Tramèr, T-RND, and color perturbation attacks, against SNNs and their ANN counterparts on MNIST, Fashion-MNIST, and CIFAR-10","url":"https://arxiv.org/abs/2610.36298","relationship":"Source-stated"},{"text":"The authors report that SNNs were consistently harder to steal across every tested attack and dataset; SNN surrogates had lower agreement with target models and lower adversarial transferability than ANN surrogates.","excerpt":"Across every attack and dataset, SNNs prove consistently more difficult to steal. Surrogates of SNNs achieve 10% to 20% lower agreement with the target model and 15% to 25% lower adversarial transferability than surrogates of ANNs.","url":"https://arxiv.org/abs/2610.36298","relationship":"Source-stated"},{"text":"The authors attribute the reported resistance to SNNs’ spike-based, time-dependent dynamics and argue that the findings strengthen the case for neuromorphic architectures in security-sensitive applications.","excerpt":"These findings extend the known resilience of SNNs beyond attacks on training data to the replication of model functionality, strengthening the case for neuromorphic architectures in security-sensitive applications.","url":"https://arxiv.org/abs/2610.36298","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2610.36298","checkedAt":"2026-11-07T08:50:08.274Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"Would the reported gap persist against other model stealing attacks or on tasks beyond the three datasets tested?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"Moshruba and Parsa argue spiking neural networks are harder to steal","summary":"Ayana Moshruba and Maryam Parsa report a benchmark study comparing model stealing attacks against spiking neural networks (SNNs) and conventional artificial neural networks (ANNs). Model stealing attacks use a model’s responses to queries to train a surrogate model that imitates its decisions, without access to its parameters or training data.\n\nThe authors tested four attacks on MNIST, Fashion-MNIST, and CIFAR-10. They report that, across every attack and dataset, SNN surrogates had 10% to 20% lower agreement with the target and 15% to 25% lower adversarial transferability than ANN surrogates. They attribute the difference to SNNs’ sparse, discrete spikes and time-dependent dynamics, which they say make their decision boundaries harder for ANN-based surrogates to approximate, even with large query budgets.\n\nThe authors argue that this resistance extends SNNs’ known resilience beyond attacks on training data to attempts to replicate model functionality, strengthening the case for neuromorphic architectures in security-sensitive applications. The abstract describes benchmark results, not how well this resistance holds in deployed systems. Would the reported gap persist against other model stealing attacks or on tasks beyond the three datasets tested?","question":"Would the reported gap persist against other model stealing attacks or on tasks beyond the three datasets tested?","claims":[{"text":"Moshruba and Parsa define model stealing attacks as using a target model’s responses to train a surrogate that replicates its decisions, without access to its parameters or training data.","excerpt":"training a surrogate model that replicates the decision behavior of a target through black-box queries alone, without access to its parameters or training data","support":"source-stated"},{"text":"The authors benchmarked four representative attacks against SNNs and their ANN counterparts on MNIST, Fashion-MNIST, and CIFAR-10.","excerpt":"benchmarking four representative MSAs, namely the Papernot, Tramèr, T-RND, and color perturbation attacks, against SNNs and their ANN counterparts on MNIST, Fashion-MNIST, and CIFAR-10","support":"source-stated"},{"text":"The authors report that SNNs were consistently harder to steal across every tested attack and dataset; SNN surrogates had lower agreement with target models and lower adversarial transferability than ANN surrogates.","excerpt":"Across every attack and dataset, SNNs prove consistently more difficult to steal. Surrogates of SNNs achieve 10% to 20% lower agreement with the target model and 15% to 25% lower adversarial transferability than surrogates of ANNs.","support":"source-stated"},{"text":"The authors attribute the reported resistance to SNNs’ spike-based, time-dependent dynamics and argue that the findings strengthen the case for neuromorphic architectures in security-sensitive applications.","excerpt":"These findings extend the known resilience of SNNs beyond attacks on training data to the replication of model functionality, strengthening the case for neuromorphic architectures in security-sensitive applications.","support":"source-stated"}],"limitations":["The evidence supplied is the arXiv abstract; it does not provide details of the experimental setup beyond the named attacks and datasets.","The reported findings concern four attacks and three datasets; the abstract does not establish whether the results generalize to other attacks, tasks, or deployed systems.","The authors attribute the resistance to SNN computation, but the supplied abstract does not provide further evidence or analysis of that explanation."],"mediaId":"0ed6edd9404f35cfd939b968b2e1975a24c3d37297797bb757c808dd261574b7","correctionReason":null,"id":"f15db2cf472f0b25310408ba33c84fc679190b09ae9e0fe4fa97257cd4fc25a0","publicationId":"1b35daa8-a32f-48cf-82ec-60c86fb883c2","version":1,"sourceUrl":"https://arxiv.org/abs/2610.36298","sourceTitle":"Spiking Neural Networks Are Harder to Steal: A Benchmark Study of Model Stealing Attacks","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-10-26","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-10-28T07:20:36.406Z","createdAt":"2026-11-07T07:47:42.813Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"0ed6edd9404f35cfd939b968b2e1975a24c3d37297797bb757c808dd261574b7","kind":"article","url":"https://arxiv.org/abs/2610.36298","sourceUrl":"https://arxiv.org/abs/2610.36298","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-10-30T23:03:35.276Z","method":"acquired_source","contentType":"text/html"},"recommendationReason":"Recommended by Jev for understanding this source; checked by Luna."}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2610.36298","checkedAt":"2026-11-07T08:50:08.274Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v5","mediaId":"0ed6edd9404f35cfd939b968b2e1975a24c3d37297797bb757c808dd261574b7","confidence":0.84,"provider":"typesafe","model":"jev-1.13.0","inputHash":"ea79bef61a080c0287518380779fe07cb14df967863aaa20fb255e3757e48f83","decidedAt":"2026-10-30T23:03:37.223Z"},"provenance":{"textHash":"c7a95ba7227ab369886d6c80ed0b0bcec57c8d5303355b7b28bf701781be9f48","observationId":"8090a367531858ec220485f91926b24e262c499772a8093cd000142add35bc14","retrievedAt":"2026-10-30T23:03:34.971Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"b5725063-5600-4373-81a5-b657cb2cbf57"},"reviewer":{"model":"gpt-6-luna","runId":"ffb20c35-728a-458e-89c6-de0e2f73e887"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-07T07:47:42.813Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Moshruba, Ayana","url":null,"kind":"unknown"},{"name":"Parsa, Maryam","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Moshruba, Ayana","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Moshruba, Ayana","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.36298","locator":"head/meta[name=author|citation_author]"},"assertionId":"297b3dac5c900fe7a2b0a91e19b0a0075b9637d88a2a44bc34ebf61f5a1c14c7:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-07T08:50:08.274Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"297b3dac5c900fe7a2b0a91e19b0a0075b9637d88a2a44bc34ebf61f5a1c14c7","sourceUrl":"https://arxiv.org/abs/2610.36298","contentSha256":"941b9a0193fc398415c104ae9ce9af057b6658a1a327d5cd448ab771de8cf7ad","textHash":"c7a95ba7227ab369886d6c80ed0b0bcec57c8d5303355b7b28bf701781be9f48","observedAt":"2026-11-07T08:50:08.274Z"}},{"actor":{"name":"Parsa, Maryam","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Parsa, Maryam","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.36298","locator":"head/meta[name=author|citation_author]"},"assertionId":"297b3dac5c900fe7a2b0a91e19b0a0075b9637d88a2a44bc34ebf61f5a1c14c7:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-07T08:50:08.274Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"297b3dac5c900fe7a2b0a91e19b0a0075b9637d88a2a44bc34ebf61f5a1c14c7","sourceUrl":"https://arxiv.org/abs/2610.36298","contentSha256":"941b9a0193fc398415c104ae9ce9af057b6658a1a327d5cd448ab771de8cf7ad","textHash":"c7a95ba7227ab369886d6c80ed0b0bcec57c8d5303355b7b28bf701781be9f48","observedAt":"2026-11-07T08:50:08.274Z"}}],"participationCoverage":{"total":2,"returned":2,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2610.36298","snapshotId":"297b3dac5c900fe7a2b0a91e19b0a0075b9637d88a2a44bc34ebf61f5a1c14c7","contentSha256":"941b9a0193fc398415c104ae9ce9af057b6658a1a327d5cd448ab771de8cf7ad","textHash":"c7a95ba7227ab369886d6c80ed0b0bcec57c8d5303355b7b28bf701781be9f48","observedAt":"2026-11-07T08:50:08.274Z","basis":"publisher-page-metadata","sourceLinks":40,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/f15db2cf472f0b25310408ba33c84fc679190b09ae9e0fe4fa97257cd4fc25a0/source"}}}],"audits":[{"publicationId":"1b35daa8-a32f-48cf-82ec-60c86fb883c2","verdict":"uncertain","findings":[{"dimension":"attribution","status":"pass","explanation":"The reading correctly credits the paper to Ayana Moshruba and Maryam Parsa. It keeps the results, the causal explanation and the security implication framed as the authors' report and argument.","excerpt":"Authors:Ayana Moshruba, Maryam Parsa"},{"dimension":"summary","status":"uncertain","explanation":"The reading is faithful, but central terms are left unexplained for a non-specialist. It never plainly says what an SNN is: brain-inspired networks that send discrete spikes whose outputs build up over time. Adversarial transferability, one of only two headline metrics, is not explained, and neither is neuromorphic architectures. SNN surrogates could also be misread as surrogates built from SNNs. Smallest repairs: add a one-clause SNN definition, gloss adversarial transferability, and write surrogates of SNNs, matching the source.","excerpt":"SNNs communicate through sparse, discrete spikes, and their outputs depend on membrane potentials that evolve over time rather than on a static mapping from input to output."},{"dimension":"date","status":"pass","explanation":"The original publication date, 2026-10-26, matches the v1 submission date. The later feed date of 2026-10-29 was correctly not used.","excerpt":"[Submitted on 26 Oct 2026]"},{"dimension":"support","status":"pass","explanation":"The figures, attacks, datasets, causal attribution and security claim all match the abstract. The caveat that these are benchmark results rather than evidence about deployed systems is properly labeled. The closing question follows from the stated scope and is not a leading claim.","excerpt":"Surrogates of SNNs achieve 10% to 20% lower agreement with the target model and 15% to 25% lower adversarial transferability than surrogates of ANNs."},{"dimension":"media","status":"pass","explanation":"The selected media is the verified arXiv abstract page for this paper. It is the direct original source.","excerpt":null}],"createdAt":"2026-11-07T07:47:42.813Z","model":"anthropic/claude-opus-5.5"}],"correctionUrl":"https://escapemeridian.com/corrections/"}