{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:KSC3PADBWYUI5ERDL532GW5E3N","short_pith_number":"pith:KSC3PADB","canonical_record":{"source":{"id":"2306.06098","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-06-09T17:58:47Z","cross_cats_sorted":["cs.NA","math.NA","math.OC"],"title_canon_sha256":"293a74fedda2b58a815ed624f5b13922d3fb749c2b22ebfa4b47750d3c74a535","abstract_canon_sha256":"a78a0407e816e48edc2fa7efbe3933b554946d139cc88dbfb88b8e3f053528b8"},"schema_version":"1.0"},"canonical_sha256":"5485b78061b6288e92235f77a35ba4db685391c92c25aee8c35fd00341432066","source":{"kind":"arxiv","id":"2306.06098","version":5},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.06098","created_at":"2026-07-05T08:27:40Z"},{"alias_kind":"arxiv_version","alias_value":"2306.06098v5","created_at":"2026-07-05T08:27:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.06098","created_at":"2026-07-05T08:27:40Z"},{"alias_kind":"pith_short_12","alias_value":"KSC3PADBWYUI","created_at":"2026-07-05T08:27:40Z"},{"alias_kind":"pith_short_16","alias_value":"KSC3PADBWYUI5ERD","created_at":"2026-07-05T08:27:40Z"},{"alias_kind":"pith_short_8","alias_value":"KSC3PADB","created_at":"2026-07-05T08:27:40Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:KSC3PADBWYUI5ERDL532GW5E3N","target":"record","payload":{"canonical_record":{"source":{"id":"2306.06098","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-06-09T17:58:47Z","cross_cats_sorted":["cs.NA","math.NA","math.OC"],"title_canon_sha256":"293a74fedda2b58a815ed624f5b13922d3fb749c2b22ebfa4b47750d3c74a535","abstract_canon_sha256":"a78a0407e816e48edc2fa7efbe3933b554946d139cc88dbfb88b8e3f053528b8"},"schema_version":"1.0"},"canonical_sha256":"5485b78061b6288e92235f77a35ba4db685391c92c25aee8c35fd00341432066","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:27:40.732531Z","signature_b64":"rs5nfuQQorpARpOHeB0JXWVZqOPClErK7iXrNMLBedP0PTSnUzpDtDhrxp2igTKY9Vjpdhwc6Ca87zWs+gzTCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5485b78061b6288e92235f77a35ba4db685391c92c25aee8c35fd00341432066","last_reissued_at":"2026-07-05T08:27:40.731946Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:27:40.731946Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2306.06098","source_version":5,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:27:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"piqstXCgkc/TkN19E20Jpifk9TKeBaYyM6VY5sOZU65g5fhM+vKUG1QCnHjBNy4LzI6V4mWJwhlRbeJ3Uhq8Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T08:01:57.065885Z"},"content_sha256":"6584a20089d7ce2bb6ededa770c9236ec2611ad7b040be571f8ed6838e8b34d3","schema_version":"1.0","event_id":"sha256:6584a20089d7ce2bb6ededa770c9236ec2611ad7b040be571f8ed6838e8b34d3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:KSC3PADBWYUI5ERDL532GW5E3N","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Error Feedback Can Accurately Compress Preconditioners","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.NA","math.NA","math.OC"],"primary_cat":"cs.LG","authors_text":"Aleksei Kalinov, Dan Alistarh, Eldar Kurtic, Elias Frantar, Ionut-Vlad Modoranu","submitted_at":"2023-06-09T17:58:47Z","abstract_excerpt":"Leveraging second-order information about the loss at the scale of deep networks is one of the main lines of approach for improving the performance of current optimizers for deep learning. Yet, existing approaches for accurate full-matrix preconditioning, such as Full-Matrix Adagrad (GGT) or Matrix-Free Approximate Curvature (M-FAC) suffer from massive storage costs when applied even to small-scale models, as they must store a sliding window of gradients, whose memory requirements are multiplicative in the model dimension. In this paper, we address this issue via a novel and efficient error-fe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.06098","kind":"arxiv","version":5},"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/2306.06098/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":"2026-07-05T08:27:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Qegx+NdhglQbcRgai1S9N0oSTuxFtABNqISKxN/jGdlhYt5Rs5pF6kLr+ugrEBFxwz/RNT5Q+oPtta1Ci4ClBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T08:01:57.066268Z"},"content_sha256":"7d698a02103415c081d644dcf04ad59e00195e2651cb84de74ef291326f624b2","schema_version":"1.0","event_id":"sha256:7d698a02103415c081d644dcf04ad59e00195e2651cb84de74ef291326f624b2"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KSC3PADBWYUI5ERDL532GW5E3N/bundle.json","state_url":"https://pith.science/pith/KSC3PADBWYUI5ERDL532GW5E3N/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KSC3PADBWYUI5ERDL532GW5E3N/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":"2026-07-07T08:01:57Z","links":{"resolver":"https://pith.science/pith/KSC3PADBWYUI5ERDL532GW5E3N","bundle":"https://pith.science/pith/KSC3PADBWYUI5ERDL532GW5E3N/bundle.json","state":"https://pith.science/pith/KSC3PADBWYUI5ERDL532GW5E3N/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KSC3PADBWYUI5ERDL532GW5E3N/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:KSC3PADBWYUI5ERDL532GW5E3N","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":"a78a0407e816e48edc2fa7efbe3933b554946d139cc88dbfb88b8e3f053528b8","cross_cats_sorted":["cs.NA","math.NA","math.OC"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-06-09T17:58:47Z","title_canon_sha256":"293a74fedda2b58a815ed624f5b13922d3fb749c2b22ebfa4b47750d3c74a535"},"schema_version":"1.0","source":{"id":"2306.06098","kind":"arxiv","version":5}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.06098","created_at":"2026-07-05T08:27:40Z"},{"alias_kind":"arxiv_version","alias_value":"2306.06098v5","created_at":"2026-07-05T08:27:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.06098","created_at":"2026-07-05T08:27:40Z"},{"alias_kind":"pith_short_12","alias_value":"KSC3PADBWYUI","created_at":"2026-07-05T08:27:40Z"},{"alias_kind":"pith_short_16","alias_value":"KSC3PADBWYUI5ERD","created_at":"2026-07-05T08:27:40Z"},{"alias_kind":"pith_short_8","alias_value":"KSC3PADB","created_at":"2026-07-05T08:27:40Z"}],"graph_snapshots":[{"event_id":"sha256:7d698a02103415c081d644dcf04ad59e00195e2651cb84de74ef291326f624b2","target":"graph","created_at":"2026-07-05T08:27:40Z","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/2306.06098/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Leveraging second-order information about the loss at the scale of deep networks is one of the main lines of approach for improving the performance of current optimizers for deep learning. Yet, existing approaches for accurate full-matrix preconditioning, such as Full-Matrix Adagrad (GGT) or Matrix-Free Approximate Curvature (M-FAC) suffer from massive storage costs when applied even to small-scale models, as they must store a sliding window of gradients, whose memory requirements are multiplicative in the model dimension. In this paper, we address this issue via a novel and efficient error-fe","authors_text":"Aleksei Kalinov, Dan Alistarh, Eldar Kurtic, Elias Frantar, Ionut-Vlad Modoranu","cross_cats":["cs.NA","math.NA","math.OC"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-06-09T17:58:47Z","title":"Error Feedback Can Accurately Compress Preconditioners"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.06098","kind":"arxiv","version":5},"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":"sha256:6584a20089d7ce2bb6ededa770c9236ec2611ad7b040be571f8ed6838e8b34d3","target":"record","created_at":"2026-07-05T08:27:40Z","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":"a78a0407e816e48edc2fa7efbe3933b554946d139cc88dbfb88b8e3f053528b8","cross_cats_sorted":["cs.NA","math.NA","math.OC"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-06-09T17:58:47Z","title_canon_sha256":"293a74fedda2b58a815ed624f5b13922d3fb749c2b22ebfa4b47750d3c74a535"},"schema_version":"1.0","source":{"id":"2306.06098","kind":"arxiv","version":5}},"canonical_sha256":"5485b78061b6288e92235f77a35ba4db685391c92c25aee8c35fd00341432066","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5485b78061b6288e92235f77a35ba4db685391c92c25aee8c35fd00341432066","first_computed_at":"2026-07-05T08:27:40.731946Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:27:40.731946Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"rs5nfuQQorpARpOHeB0JXWVZqOPClErK7iXrNMLBedP0PTSnUzpDtDhrxp2igTKY9Vjpdhwc6Ca87zWs+gzTCg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:27:40.732531Z","signed_message":"canonical_sha256_bytes"},"source_id":"2306.06098","source_kind":"arxiv","source_version":5}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6584a20089d7ce2bb6ededa770c9236ec2611ad7b040be571f8ed6838e8b34d3","sha256:7d698a02103415c081d644dcf04ad59e00195e2651cb84de74ef291326f624b2"],"state_sha256":"0205520771efe3b65115dd22512050d630fb98493e08bd1632b24769a047781e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wyUtArrdE5m4Ys/E9XKIOB/3rkZbzr7IfpGtHOLPKk/PKIfz1Vnzj8qQcwLICa68Y9x+Dauf0EfpTlzLsMsOCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T08:01:57.068425Z","bundle_sha256":"8b99ab302e648aa00c1b8bd574ec7a55c47964a0f4e098408cd5ac1fb384fbf3"}}