{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:YDK4SF6TNZE6N3HVM4SEXI3777","short_pith_number":"pith:YDK4SF6T","canonical_record":{"source":{"id":"2605.08022","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.NE","submitted_at":"2026-05-08T17:10:08Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"5270f8f000bddbd66d42c49a0d160ac2f7e7ce826c037fbdb5840ca9bb242d07","abstract_canon_sha256":"62dbc7deb161d1d3193e6963e64c2778b78d1752a69e060defc178d54adde81a"},"schema_version":"1.0"},"canonical_sha256":"c0d5c917d36e49e6ecf567244ba37fffce8c564cc0450d4538645e84e34e2ccc","source":{"kind":"arxiv","id":"2605.08022","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.08022","created_at":"2026-06-30T00:15:09Z"},{"alias_kind":"arxiv_version","alias_value":"2605.08022v2","created_at":"2026-06-30T00:15:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.08022","created_at":"2026-06-30T00:15:09Z"},{"alias_kind":"pith_short_12","alias_value":"YDK4SF6TNZE6","created_at":"2026-06-30T00:15:09Z"},{"alias_kind":"pith_short_16","alias_value":"YDK4SF6TNZE6N3HV","created_at":"2026-06-30T00:15:09Z"},{"alias_kind":"pith_short_8","alias_value":"YDK4SF6T","created_at":"2026-06-30T00:15:09Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:YDK4SF6TNZE6N3HVM4SEXI3777","target":"record","payload":{"canonical_record":{"source":{"id":"2605.08022","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.NE","submitted_at":"2026-05-08T17:10:08Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"5270f8f000bddbd66d42c49a0d160ac2f7e7ce826c037fbdb5840ca9bb242d07","abstract_canon_sha256":"62dbc7deb161d1d3193e6963e64c2778b78d1752a69e060defc178d54adde81a"},"schema_version":"1.0"},"canonical_sha256":"c0d5c917d36e49e6ecf567244ba37fffce8c564cc0450d4538645e84e34e2ccc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T00:15:09.898046Z","signature_b64":"sHJ5UWpj9QM21UihDNDj11yu+64cdRk/S4uB7aj8ZL8vJn7ppJ9ZiWXJlwUjiXuTsNBoiK3oxgQjNsgV2R4mAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c0d5c917d36e49e6ecf567244ba37fffce8c564cc0450d4538645e84e34e2ccc","last_reissued_at":"2026-06-30T00:15:09.897576Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T00:15:09.897576Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.08022","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-30T00:15:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rMNMChe99FbCi0pSBotsT5kmbsjyaceX/sSlaO9NkY0Pf9CEF8R7nDn07GcabNuE41tHDQUCLAJwdmxlQFKFBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T11:56:35.586565Z"},"content_sha256":"f74725898b8333cd9c3f5ba04c105eb189915232b620ee059948616ad19a0650","schema_version":"1.0","event_id":"sha256:f74725898b8333cd9c3f5ba04c105eb189915232b620ee059948616ad19a0650"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:YDK4SF6TNZE6N3HVM4SEXI3777","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Globally Optimal Training of Spiking Neural Networks via Parameter Reconstruction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Extending convexification to recurrent threshold networks enables a parameter reconstruction algorithm for globally optimal SNN training.","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.NE","authors_text":"ChengXiang Zhai, Himanshu Udupi, Xiaocong Yang","submitted_at":"2026-05-08T17:10:08Z","abstract_excerpt":"Spiking Neural Networks (SNNs) have been proposed as biologically plausible and energy-efficient alternatives to conventional Artificial Neural Networks (ANNs). However, the training of SNN usually relies on surrogate gradients due to the non-differentiability of the spike function, introducing approximation errors that accumulate across layers. To address this challenge, we extend the work on convexification of parallel feedforward threshold networks to parallel recurrent threshold networks, which subsume parallel SNNs as a structured special case. Building on this theoretical framework, we p"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"we propose a parameter reconstruction algorithm for SNN training that demonstrates consistent and significant advantages across various tasks, both as a standalone method and in combination with surrogate-gradient training.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That extending convexification from parallel feedforward threshold networks to parallel recurrent threshold networks is valid and that this subsumes SNNs as a structured special case allowing global optimality via parameter reconstruction.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A new parameter reconstruction method achieves globally optimal training for spiking neural networks by convexifying parallel recurrent threshold networks that include SNNs as a special case.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Extending convexification to recurrent threshold networks enables a parameter reconstruction algorithm for globally optimal SNN training.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"7cf53377db225afe96faaacc409947e1efa169cbaba8456e6bb595b076175756"},"source":{"id":"2605.08022","kind":"arxiv","version":2},"verdict":{"id":"7c484d44-7265-49aa-b4cb-258087fc6fa0","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-11T02:15:00.961955Z","strongest_claim":"we propose a parameter reconstruction algorithm for SNN training that demonstrates consistent and significant advantages across various tasks, both as a standalone method and in combination with surrogate-gradient training.","one_line_summary":"A new parameter reconstruction method achieves globally optimal training for spiking neural networks by convexifying parallel recurrent threshold networks that include SNNs as a special case.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That extending convexification from parallel feedforward threshold networks to parallel recurrent threshold networks is valid and that this subsumes SNNs as a structured special case allowing global optimality via parameter reconstruction.","pith_extraction_headline":"Extending convexification to recurrent threshold networks enables a parameter reconstruction algorithm for globally optimal SNN training."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.08022/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T09:42:04.560083Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-20T04:42:55.281353Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T15:01:18.647153Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T11:18:38.443444Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"58f26156e18ad356c4308c41b4e011307bb0ac5499dd2d784adb83964995efbb"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"7627c6faa1af2ec8e022e6cb276c18338dfe2a00dc84e300a2821c62947a4aa1"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"7c484d44-7265-49aa-b4cb-258087fc6fa0"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-30T00:15:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DpIkHELKGs22XQt9iSIuNxrDkq6mAKnVvdVaySnzlsXuAbkz7vU4yEu2m1CIBQGTEXiXhsgIHxVaTt3HWIjvCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T11:56:35.587285Z"},"content_sha256":"4f237f6a6c0277d6de07ecb1e3da0f9e0ffb0454e30659248d9752f0565c22d1","schema_version":"1.0","event_id":"sha256:4f237f6a6c0277d6de07ecb1e3da0f9e0ffb0454e30659248d9752f0565c22d1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YDK4SF6TNZE6N3HVM4SEXI3777/bundle.json","state_url":"https://pith.science/pith/YDK4SF6TNZE6N3HVM4SEXI3777/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YDK4SF6TNZE6N3HVM4SEXI3777/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-15T11:56:35Z","links":{"resolver":"https://pith.science/pith/YDK4SF6TNZE6N3HVM4SEXI3777","bundle":"https://pith.science/pith/YDK4SF6TNZE6N3HVM4SEXI3777/bundle.json","state":"https://pith.science/pith/YDK4SF6TNZE6N3HVM4SEXI3777/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YDK4SF6TNZE6N3HVM4SEXI3777/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:YDK4SF6TNZE6N3HVM4SEXI3777","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"62dbc7deb161d1d3193e6963e64c2778b78d1752a69e060defc178d54adde81a","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.NE","submitted_at":"2026-05-08T17:10:08Z","title_canon_sha256":"5270f8f000bddbd66d42c49a0d160ac2f7e7ce826c037fbdb5840ca9bb242d07"},"schema_version":"1.0","source":{"id":"2605.08022","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.08022","created_at":"2026-06-30T00:15:09Z"},{"alias_kind":"arxiv_version","alias_value":"2605.08022v2","created_at":"2026-06-30T00:15:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.08022","created_at":"2026-06-30T00:15:09Z"},{"alias_kind":"pith_short_12","alias_value":"YDK4SF6TNZE6","created_at":"2026-06-30T00:15:09Z"},{"alias_kind":"pith_short_16","alias_value":"YDK4SF6TNZE6N3HV","created_at":"2026-06-30T00:15:09Z"},{"alias_kind":"pith_short_8","alias_value":"YDK4SF6T","created_at":"2026-06-30T00:15:09Z"}],"graph_snapshots":[{"event_id":"sha256:4f237f6a6c0277d6de07ecb1e3da0f9e0ffb0454e30659248d9752f0565c22d1","target":"graph","created_at":"2026-06-30T00:15:09Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"we propose a parameter reconstruction algorithm for SNN training that demonstrates consistent and significant advantages across various tasks, both as a standalone method and in combination with surrogate-gradient training."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That extending convexification from parallel feedforward threshold networks to parallel recurrent threshold networks is valid and that this subsumes SNNs as a structured special case allowing global optimality via parameter reconstruction."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"A new parameter reconstruction method achieves globally optimal training for spiking neural networks by convexifying parallel recurrent threshold networks that include SNNs as a special case."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Extending convexification to recurrent threshold networks enables a parameter reconstruction algorithm for globally optimal SNN training."}],"snapshot_sha256":"7cf53377db225afe96faaacc409947e1efa169cbaba8456e6bb595b076175756"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"7627c6faa1af2ec8e022e6cb276c18338dfe2a00dc84e300a2821c62947a4aa1"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-20T09:42:04.560083Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-20T04:42:55.281353Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T15:01:18.647153Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T11:18:38.443444Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.08022/integrity.json","findings":[],"snapshot_sha256":"58f26156e18ad356c4308c41b4e011307bb0ac5499dd2d784adb83964995efbb","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Spiking Neural Networks (SNNs) have been proposed as biologically plausible and energy-efficient alternatives to conventional Artificial Neural Networks (ANNs). However, the training of SNN usually relies on surrogate gradients due to the non-differentiability of the spike function, introducing approximation errors that accumulate across layers. To address this challenge, we extend the work on convexification of parallel feedforward threshold networks to parallel recurrent threshold networks, which subsume parallel SNNs as a structured special case. Building on this theoretical framework, we p","authors_text":"ChengXiang Zhai, Himanshu Udupi, Xiaocong Yang","cross_cats":["cs.AI","cs.LG"],"headline":"Extending convexification to recurrent threshold networks enables a parameter reconstruction algorithm for globally optimal SNN training.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.NE","submitted_at":"2026-05-08T17:10:08Z","title":"Globally Optimal Training of Spiking Neural Networks via Parameter Reconstruction"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.08022","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-11T02:15:00.961955Z","id":"7c484d44-7265-49aa-b4cb-258087fc6fa0","model_set":{"reader":"grok-4.3"},"one_line_summary":"A new parameter reconstruction method achieves globally optimal training for spiking neural networks by convexifying parallel recurrent threshold networks that include SNNs as a special case.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Extending convexification to recurrent threshold networks enables a parameter reconstruction algorithm for globally optimal SNN training.","strongest_claim":"we propose a parameter reconstruction algorithm for SNN training that demonstrates consistent and significant advantages across various tasks, both as a standalone method and in combination with surrogate-gradient training.","weakest_assumption":"That extending convexification from parallel feedforward threshold networks to parallel recurrent threshold networks is valid and that this subsumes SNNs as a structured special case allowing global optimality via parameter reconstruction."}},"verdict_id":"7c484d44-7265-49aa-b4cb-258087fc6fa0"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:f74725898b8333cd9c3f5ba04c105eb189915232b620ee059948616ad19a0650","target":"record","created_at":"2026-06-30T00:15:09Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"62dbc7deb161d1d3193e6963e64c2778b78d1752a69e060defc178d54adde81a","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.NE","submitted_at":"2026-05-08T17:10:08Z","title_canon_sha256":"5270f8f000bddbd66d42c49a0d160ac2f7e7ce826c037fbdb5840ca9bb242d07"},"schema_version":"1.0","source":{"id":"2605.08022","kind":"arxiv","version":2}},"canonical_sha256":"c0d5c917d36e49e6ecf567244ba37fffce8c564cc0450d4538645e84e34e2ccc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c0d5c917d36e49e6ecf567244ba37fffce8c564cc0450d4538645e84e34e2ccc","first_computed_at":"2026-06-30T00:15:09.897576Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-30T00:15:09.897576Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"sHJ5UWpj9QM21UihDNDj11yu+64cdRk/S4uB7aj8ZL8vJn7ppJ9ZiWXJlwUjiXuTsNBoiK3oxgQjNsgV2R4mAg==","signature_status":"signed_v1","signed_at":"2026-06-30T00:15:09.898046Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.08022","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f74725898b8333cd9c3f5ba04c105eb189915232b620ee059948616ad19a0650","sha256:4f237f6a6c0277d6de07ecb1e3da0f9e0ffb0454e30659248d9752f0565c22d1"],"state_sha256":"00c9f18ba68affd10ab6d00b78bbe1b17752486158027240361c43147ad257f8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jH/iRtg1m/44Nf+BU40fawi1+JN+isFqYgttEwuorGz3FhnA+wMs/iDI3jBr4cfWfSPDqetALNQrtq7eP8OjAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-15T11:56:35.591588Z","bundle_sha256":"d2dab5e393494e118f195cefd0a0f2c49779b72420c8d56068c46cd4695991a2"}}