{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:ZSOC4RONXU44NLD2D2ETPFQCGV","short_pith_number":"pith:ZSOC4RON","canonical_record":{"source":{"id":"2505.17962","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-23T14:33:20Z","cross_cats_sorted":[],"title_canon_sha256":"f95d1353b79eb4609f0224cd1991a38d1c3efe0f082e9271f74ffe692b7a6151","abstract_canon_sha256":"99d42abdc1f51a9234b06dc1db2e0d9bb9ae0daee6b09c66fbc00684809a68ab"},"schema_version":"1.0"},"canonical_sha256":"cc9c2e45cdbd39c6ac7a1e89379602355706dec650cbcb481096887945befd60","source":{"kind":"arxiv","id":"2505.17962","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.17962","created_at":"2026-07-05T11:08:34Z"},{"alias_kind":"arxiv_version","alias_value":"2505.17962v1","created_at":"2026-07-05T11:08:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.17962","created_at":"2026-07-05T11:08:34Z"},{"alias_kind":"pith_short_12","alias_value":"ZSOC4RONXU44","created_at":"2026-07-05T11:08:34Z"},{"alias_kind":"pith_short_16","alias_value":"ZSOC4RONXU44NLD2","created_at":"2026-07-05T11:08:34Z"},{"alias_kind":"pith_short_8","alias_value":"ZSOC4RON","created_at":"2026-07-05T11:08:34Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:ZSOC4RONXU44NLD2D2ETPFQCGV","target":"record","payload":{"canonical_record":{"source":{"id":"2505.17962","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-23T14:33:20Z","cross_cats_sorted":[],"title_canon_sha256":"f95d1353b79eb4609f0224cd1991a38d1c3efe0f082e9271f74ffe692b7a6151","abstract_canon_sha256":"99d42abdc1f51a9234b06dc1db2e0d9bb9ae0daee6b09c66fbc00684809a68ab"},"schema_version":"1.0"},"canonical_sha256":"cc9c2e45cdbd39c6ac7a1e89379602355706dec650cbcb481096887945befd60","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:08:34.338336Z","signature_b64":"G2bDEUH8Zf6MSRfFVfoziTsxv35M2sP9/0o/3L9eit2HhWPC5iELz66rhwzeWyZ2os0FrUMz+vrn/8ADhWuHDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cc9c2e45cdbd39c6ac7a1e89379602355706dec650cbcb481096887945befd60","last_reissued_at":"2026-07-05T11:08:34.337735Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:08:34.337735Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.17962","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:08:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5Jhxrm3Q9NnTSm+Z+8/qMwVYM8lj66dJxLqB3G+Hdby/rLhMOH2GHVJ5IkK2cMmwL7N7hsIKkkmSExxWBfHhCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T22:37:53.710587Z"},"content_sha256":"b60ba2c3613b920dc9aad183b8e633ab12735b1d6f267a0b673985480865353b","schema_version":"1.0","event_id":"sha256:b60ba2c3613b920dc9aad183b8e633ab12735b1d6f267a0b673985480865353b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:ZSOC4RONXU44NLD2D2ETPFQCGV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"A Principled Bayesian Framework for Training Binary and Spiking Neural Networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Adeel Razi, James A. Walker, Moein Khajehnejad","submitted_at":"2025-05-23T14:33:20Z","abstract_excerpt":"We propose a Bayesian framework for training binary and spiking neural networks that achieves state-of-the-art performance without normalisation layers. Unlike commonly used surrogate gradient methods -- often heuristic and sensitive to hyperparameter choices -- our approach is grounded in a probabilistic model of noisy binary networks, enabling fully end-to-end gradient-based optimisation. We introduce importance-weighted straight-through (IW-ST) estimators, a unified class generalising straight-through and relaxation-based estimators. We characterise the bias-variance trade-off in this famil"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.17962","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2505.17962/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:08:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"m5pBkjzAkHEvBUqf0U0v89amyJi3DixBI9K2XgKX6DJwCxWDtPGIiZCas9X9P4foCyRaEtx+P3wgIRvUhbaXAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T22:37:53.711187Z"},"content_sha256":"64a0f6d1bda643f893fc8a9e9cf624c1835ed4b9a5ac503d1df801f9981f2217","schema_version":"1.0","event_id":"sha256:64a0f6d1bda643f893fc8a9e9cf624c1835ed4b9a5ac503d1df801f9981f2217"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZSOC4RONXU44NLD2D2ETPFQCGV/bundle.json","state_url":"https://pith.science/pith/ZSOC4RONXU44NLD2D2ETPFQCGV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZSOC4RONXU44NLD2D2ETPFQCGV/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":"2026-08-07T22:37:53Z","links":{"resolver":"https://pith.science/pith/ZSOC4RONXU44NLD2D2ETPFQCGV","bundle":"https://pith.science/pith/ZSOC4RONXU44NLD2D2ETPFQCGV/bundle.json","state":"https://pith.science/pith/ZSOC4RONXU44NLD2D2ETPFQCGV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZSOC4RONXU44NLD2D2ETPFQCGV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ZSOC4RONXU44NLD2D2ETPFQCGV","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":"99d42abdc1f51a9234b06dc1db2e0d9bb9ae0daee6b09c66fbc00684809a68ab","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-23T14:33:20Z","title_canon_sha256":"f95d1353b79eb4609f0224cd1991a38d1c3efe0f082e9271f74ffe692b7a6151"},"schema_version":"1.0","source":{"id":"2505.17962","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.17962","created_at":"2026-07-05T11:08:34Z"},{"alias_kind":"arxiv_version","alias_value":"2505.17962v1","created_at":"2026-07-05T11:08:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.17962","created_at":"2026-07-05T11:08:34Z"},{"alias_kind":"pith_short_12","alias_value":"ZSOC4RONXU44","created_at":"2026-07-05T11:08:34Z"},{"alias_kind":"pith_short_16","alias_value":"ZSOC4RONXU44NLD2","created_at":"2026-07-05T11:08:34Z"},{"alias_kind":"pith_short_8","alias_value":"ZSOC4RON","created_at":"2026-07-05T11:08:34Z"}],"graph_snapshots":[{"event_id":"sha256:64a0f6d1bda643f893fc8a9e9cf624c1835ed4b9a5ac503d1df801f9981f2217","target":"graph","created_at":"2026-07-05T11:08:34Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2505.17962/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We propose a Bayesian framework for training binary and spiking neural networks that achieves state-of-the-art performance without normalisation layers. Unlike commonly used surrogate gradient methods -- often heuristic and sensitive to hyperparameter choices -- our approach is grounded in a probabilistic model of noisy binary networks, enabling fully end-to-end gradient-based optimisation. We introduce importance-weighted straight-through (IW-ST) estimators, a unified class generalising straight-through and relaxation-based estimators. We characterise the bias-variance trade-off in this famil","authors_text":"Adeel Razi, James A. Walker, Moein Khajehnejad","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-23T14:33:20Z","title":"A Principled Bayesian Framework for Training Binary and Spiking Neural Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.17962","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:b60ba2c3613b920dc9aad183b8e633ab12735b1d6f267a0b673985480865353b","target":"record","created_at":"2026-07-05T11:08:34Z","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":"99d42abdc1f51a9234b06dc1db2e0d9bb9ae0daee6b09c66fbc00684809a68ab","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-23T14:33:20Z","title_canon_sha256":"f95d1353b79eb4609f0224cd1991a38d1c3efe0f082e9271f74ffe692b7a6151"},"schema_version":"1.0","source":{"id":"2505.17962","kind":"arxiv","version":1}},"canonical_sha256":"cc9c2e45cdbd39c6ac7a1e89379602355706dec650cbcb481096887945befd60","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cc9c2e45cdbd39c6ac7a1e89379602355706dec650cbcb481096887945befd60","first_computed_at":"2026-07-05T11:08:34.337735Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:08:34.337735Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"G2bDEUH8Zf6MSRfFVfoziTsxv35M2sP9/0o/3L9eit2HhWPC5iELz66rhwzeWyZ2os0FrUMz+vrn/8ADhWuHDA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:08:34.338336Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.17962","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b60ba2c3613b920dc9aad183b8e633ab12735b1d6f267a0b673985480865353b","sha256:64a0f6d1bda643f893fc8a9e9cf624c1835ed4b9a5ac503d1df801f9981f2217"],"state_sha256":"ec1938a7c1f17526105e6215cd148f0b0b5442f449b485c7db7f7a2bec46b412"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Tt927BWFdrQ0QO0LDBKJCS/fEgpM1juCHorZA64osCMo3HKkNSDR/da4ybsOsDmM18gbi1+6nE41tCB2x4WpAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T22:37:53.715108Z","bundle_sha256":"91f8475b40f0387b4f86184e71ff6c3ef6e37cbad77083a833695e013d885061"}}