{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:BDBY4RNLPE2FSWKRB4OLHTAW6O","short_pith_number":"pith:BDBY4RNL","canonical_record":{"source":{"id":"1902.08153","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-02-21T17:31:32Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"a597708cff5278e1a025d3b17270984276b77f7d0a58713452c05255ec61ca79","abstract_canon_sha256":"6c4a135afa3125a63648bac7bc5d70736f7ddad6b8cd7a80362af021cf72ca1c"},"schema_version":"1.0"},"canonical_sha256":"08c38e45ab79345959510f1cb3cc16f3a0cdef58b51ecc2ca1e0256a80fdcd0b","source":{"kind":"arxiv","id":"1902.08153","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1902.08153","created_at":"2026-07-05T01:00:59Z"},{"alias_kind":"arxiv_version","alias_value":"1902.08153v3","created_at":"2026-07-05T01:00:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1902.08153","created_at":"2026-07-05T01:00:59Z"},{"alias_kind":"pith_short_12","alias_value":"BDBY4RNLPE2F","created_at":"2026-07-05T01:00:59Z"},{"alias_kind":"pith_short_16","alias_value":"BDBY4RNLPE2FSWKR","created_at":"2026-07-05T01:00:59Z"},{"alias_kind":"pith_short_8","alias_value":"BDBY4RNL","created_at":"2026-07-05T01:00:59Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:BDBY4RNLPE2FSWKRB4OLHTAW6O","target":"record","payload":{"canonical_record":{"source":{"id":"1902.08153","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-02-21T17:31:32Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"a597708cff5278e1a025d3b17270984276b77f7d0a58713452c05255ec61ca79","abstract_canon_sha256":"6c4a135afa3125a63648bac7bc5d70736f7ddad6b8cd7a80362af021cf72ca1c"},"schema_version":"1.0"},"canonical_sha256":"08c38e45ab79345959510f1cb3cc16f3a0cdef58b51ecc2ca1e0256a80fdcd0b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:00:59.405559Z","signature_b64":"AlmspCp5lb0rAlaJxGdjbUQ0c6bgq34vfWRRwPnhnjYdJyU4xXHFu6pR6828xrfK0NaqpYZAhr/Sx6L+N6/+BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"08c38e45ab79345959510f1cb3cc16f3a0cdef58b51ecc2ca1e0256a80fdcd0b","last_reissued_at":"2026-07-05T01:00:59.405086Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:00:59.405086Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1902.08153","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T01:00:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Bab9cOQ3+hCRj4oVXTns42UzsGtcs3ahpQSA++ZVssXEVNjqqR8K7dwKlnXDijVdBFGGzwl0S4YyPce8F/kdAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T12:34:48.330118Z"},"content_sha256":"d2565ce754dd4ce1843631819fdf613aab885b17065818a39737b1474f88d795","schema_version":"1.0","event_id":"sha256:d2565ce754dd4ce1843631819fdf613aab885b17065818a39737b1474f88d795"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:BDBY4RNLPE2FSWKRB4OLHTAW6O","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Learned Step Size Quantization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Deepika Bablani, Dharmendra S. Modha, Jeffrey L. McKinstry, Rathinakumar Appuswamy, Steven K. Esser","submitted_at":"2019-02-21T17:31:32Z","abstract_excerpt":"Deep networks run with low precision operations at inference time offer power and space advantages over high precision alternatives, but need to overcome the challenge of maintaining high accuracy as precision decreases. Here, we present a method for training such networks, Learned Step Size Quantization, that achieves the highest accuracy to date on the ImageNet dataset when using models, from a variety of architectures, with weights and activations quantized to 2-, 3- or 4-bits of precision, and that can train 3-bit models that reach full precision baseline accuracy. Our approach builds upon"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1902.08153","kind":"arxiv","version":3},"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/1902.08153/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":"2026-07-05T01:00:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Xqzs1sqGpEwBIeF7bJhqsvHjH7tioQ/K26KdraFngIPAmB5k4TVPaxxvNrNxZzYRCRDBf6ikxW4GYWKGOG0kAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T12:34:48.330495Z"},"content_sha256":"65fcd01e9ddfc4b3a1b65eadc5002f4137bf393c85a2769801a09247b0f85599","schema_version":"1.0","event_id":"sha256:65fcd01e9ddfc4b3a1b65eadc5002f4137bf393c85a2769801a09247b0f85599"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BDBY4RNLPE2FSWKRB4OLHTAW6O/bundle.json","state_url":"https://pith.science/pith/BDBY4RNLPE2FSWKRB4OLHTAW6O/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BDBY4RNLPE2FSWKRB4OLHTAW6O/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":"2026-08-04T12:34:48Z","links":{"resolver":"https://pith.science/pith/BDBY4RNLPE2FSWKRB4OLHTAW6O","bundle":"https://pith.science/pith/BDBY4RNLPE2FSWKRB4OLHTAW6O/bundle.json","state":"https://pith.science/pith/BDBY4RNLPE2FSWKRB4OLHTAW6O/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BDBY4RNLPE2FSWKRB4OLHTAW6O/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:BDBY4RNLPE2FSWKRB4OLHTAW6O","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":"6c4a135afa3125a63648bac7bc5d70736f7ddad6b8cd7a80362af021cf72ca1c","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-02-21T17:31:32Z","title_canon_sha256":"a597708cff5278e1a025d3b17270984276b77f7d0a58713452c05255ec61ca79"},"schema_version":"1.0","source":{"id":"1902.08153","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1902.08153","created_at":"2026-07-05T01:00:59Z"},{"alias_kind":"arxiv_version","alias_value":"1902.08153v3","created_at":"2026-07-05T01:00:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1902.08153","created_at":"2026-07-05T01:00:59Z"},{"alias_kind":"pith_short_12","alias_value":"BDBY4RNLPE2F","created_at":"2026-07-05T01:00:59Z"},{"alias_kind":"pith_short_16","alias_value":"BDBY4RNLPE2FSWKR","created_at":"2026-07-05T01:00:59Z"},{"alias_kind":"pith_short_8","alias_value":"BDBY4RNL","created_at":"2026-07-05T01:00:59Z"}],"graph_snapshots":[{"event_id":"sha256:65fcd01e9ddfc4b3a1b65eadc5002f4137bf393c85a2769801a09247b0f85599","target":"graph","created_at":"2026-07-05T01:00:59Z","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/1902.08153/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Deep networks run with low precision operations at inference time offer power and space advantages over high precision alternatives, but need to overcome the challenge of maintaining high accuracy as precision decreases. Here, we present a method for training such networks, Learned Step Size Quantization, that achieves the highest accuracy to date on the ImageNet dataset when using models, from a variety of architectures, with weights and activations quantized to 2-, 3- or 4-bits of precision, and that can train 3-bit models that reach full precision baseline accuracy. Our approach builds upon","authors_text":"Deepika Bablani, Dharmendra S. Modha, Jeffrey L. McKinstry, Rathinakumar Appuswamy, Steven K. Esser","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-02-21T17:31:32Z","title":"Learned Step Size Quantization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1902.08153","kind":"arxiv","version":3},"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:d2565ce754dd4ce1843631819fdf613aab885b17065818a39737b1474f88d795","target":"record","created_at":"2026-07-05T01:00:59Z","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":"6c4a135afa3125a63648bac7bc5d70736f7ddad6b8cd7a80362af021cf72ca1c","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-02-21T17:31:32Z","title_canon_sha256":"a597708cff5278e1a025d3b17270984276b77f7d0a58713452c05255ec61ca79"},"schema_version":"1.0","source":{"id":"1902.08153","kind":"arxiv","version":3}},"canonical_sha256":"08c38e45ab79345959510f1cb3cc16f3a0cdef58b51ecc2ca1e0256a80fdcd0b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"08c38e45ab79345959510f1cb3cc16f3a0cdef58b51ecc2ca1e0256a80fdcd0b","first_computed_at":"2026-07-05T01:00:59.405086Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T01:00:59.405086Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"AlmspCp5lb0rAlaJxGdjbUQ0c6bgq34vfWRRwPnhnjYdJyU4xXHFu6pR6828xrfK0NaqpYZAhr/Sx6L+N6/+BQ==","signature_status":"signed_v1","signed_at":"2026-07-05T01:00:59.405559Z","signed_message":"canonical_sha256_bytes"},"source_id":"1902.08153","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d2565ce754dd4ce1843631819fdf613aab885b17065818a39737b1474f88d795","sha256:65fcd01e9ddfc4b3a1b65eadc5002f4137bf393c85a2769801a09247b0f85599"],"state_sha256":"d4e2d19bca03271d1cfcc0baacf81113ca56343870f8ecf957c6064d155d945f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8j7Xh/vTQB0qlUVE0UDHot3xaqVaxH/7I4ymRV+TUzCgkxx8xUm2aUop7AYR1Fz1yvmRiWcY9dhBb4J4lv89Cw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T12:34:48.333103Z","bundle_sha256":"eac833013e66c5702fa7adb4b80d2806c3b0c5b9d3dffa533109cd4c592dff78"}}