{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:BDMUG3NLCYBGLYKN6BFZ47GLNZ","short_pith_number":"pith:BDMUG3NL","canonical_record":{"source":{"id":"2301.13464","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-01-31T08:01:35Z","cross_cats_sorted":[],"title_canon_sha256":"0d8fedfde9396f4255fe8edeecc6f62ef6ed4921c9af6cddd0442b3ef3dd529e","abstract_canon_sha256":"ff0abeb488b1b87a4a3f859de46eda9128589bc5ce59cb485791ff11a690cc0c"},"schema_version":"1.0"},"canonical_sha256":"08d9436dab160265e14df04b9e7ccb6e5889876573a75c0e326e5369d4c1f245","source":{"kind":"arxiv","id":"2301.13464","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2301.13464","created_at":"2026-07-05T06:23:58Z"},{"alias_kind":"arxiv_version","alias_value":"2301.13464v2","created_at":"2026-07-05T06:23:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.13464","created_at":"2026-07-05T06:23:58Z"},{"alias_kind":"pith_short_12","alias_value":"BDMUG3NLCYBG","created_at":"2026-07-05T06:23:58Z"},{"alias_kind":"pith_short_16","alias_value":"BDMUG3NLCYBGLYKN","created_at":"2026-07-05T06:23:58Z"},{"alias_kind":"pith_short_8","alias_value":"BDMUG3NL","created_at":"2026-07-05T06:23:58Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:BDMUG3NLCYBGLYKN6BFZ47GLNZ","target":"record","payload":{"canonical_record":{"source":{"id":"2301.13464","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-01-31T08:01:35Z","cross_cats_sorted":[],"title_canon_sha256":"0d8fedfde9396f4255fe8edeecc6f62ef6ed4921c9af6cddd0442b3ef3dd529e","abstract_canon_sha256":"ff0abeb488b1b87a4a3f859de46eda9128589bc5ce59cb485791ff11a690cc0c"},"schema_version":"1.0"},"canonical_sha256":"08d9436dab160265e14df04b9e7ccb6e5889876573a75c0e326e5369d4c1f245","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:23:58.877874Z","signature_b64":"Sq87j+rtmSgzJ2EOY+oH2C7LD75ivkLsolWoxsZGmiiLPdLThcyqeM2nOf0gKybSaWRONw9jPXE/NE5Oh6zKAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"08d9436dab160265e14df04b9e7ccb6e5889876573a75c0e326e5369d4c1f245","last_reissued_at":"2026-07-05T06:23:58.877392Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:23:58.877392Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2301.13464","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":"2026-07-05T06:23:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"U0SAe+RoOX327tvzNYwPC/c/x57GAkM8FqkFi0KDUedrlkTc+5g3RJLoAfaj/sXY6EOm3iBqMQfY/4hLQA43Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T16:10:41.947608Z"},"content_sha256":"19b35d45cc8112701ad91c54bf31aa62f4746f1c557f0c642ee777744ce69e83","schema_version":"1.0","event_id":"sha256:19b35d45cc8112701ad91c54bf31aa62f4746f1c557f0c642ee777744ce69e83"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:BDMUG3NLCYBGLYKN6BFZ47GLNZ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Training with Mixed-Precision Floating-Point Assignments","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alex Aiken, Rahul Sharma, Wonyeol Lee","submitted_at":"2023-01-31T08:01:35Z","abstract_excerpt":"When training deep neural networks, keeping all tensors in high precision (e.g., 32-bit or even 16-bit floats) is often wasteful. However, keeping all tensors in low precision (e.g., 8-bit floats) can lead to unacceptable accuracy loss. Hence, it is important to use a precision assignment -- a mapping from all tensors (arising in training) to precision levels (high or low) -- that keeps most of the tensors in low precision and leads to sufficiently accurate models. We provide a technique that explores this memory-accuracy tradeoff by generating precision assignments for convolutional neural ne"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.13464","kind":"arxiv","version":2},"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/2301.13464/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":"2026-07-05T06:23:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"g1Zx/FuloHyBImG4H9fb46F0D45Y8cSrUPmXOCXLl5a+IcCH8Bx3VnR5XopMvKc9TyXl/ndIMwS2TpKQqmgmCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T16:10:41.948163Z"},"content_sha256":"908e5d345d05beb999119edadac60c4c9efb3f7f6a815e6315b2e7672fdf4ce3","schema_version":"1.0","event_id":"sha256:908e5d345d05beb999119edadac60c4c9efb3f7f6a815e6315b2e7672fdf4ce3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BDMUG3NLCYBGLYKN6BFZ47GLNZ/bundle.json","state_url":"https://pith.science/pith/BDMUG3NLCYBGLYKN6BFZ47GLNZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BDMUG3NLCYBGLYKN6BFZ47GLNZ/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":"2026-08-19T16:10:41Z","links":{"resolver":"https://pith.science/pith/BDMUG3NLCYBGLYKN6BFZ47GLNZ","bundle":"https://pith.science/pith/BDMUG3NLCYBGLYKN6BFZ47GLNZ/bundle.json","state":"https://pith.science/pith/BDMUG3NLCYBGLYKN6BFZ47GLNZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BDMUG3NLCYBGLYKN6BFZ47GLNZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:BDMUG3NLCYBGLYKN6BFZ47GLNZ","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":"ff0abeb488b1b87a4a3f859de46eda9128589bc5ce59cb485791ff11a690cc0c","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-01-31T08:01:35Z","title_canon_sha256":"0d8fedfde9396f4255fe8edeecc6f62ef6ed4921c9af6cddd0442b3ef3dd529e"},"schema_version":"1.0","source":{"id":"2301.13464","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2301.13464","created_at":"2026-07-05T06:23:58Z"},{"alias_kind":"arxiv_version","alias_value":"2301.13464v2","created_at":"2026-07-05T06:23:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.13464","created_at":"2026-07-05T06:23:58Z"},{"alias_kind":"pith_short_12","alias_value":"BDMUG3NLCYBG","created_at":"2026-07-05T06:23:58Z"},{"alias_kind":"pith_short_16","alias_value":"BDMUG3NLCYBGLYKN","created_at":"2026-07-05T06:23:58Z"},{"alias_kind":"pith_short_8","alias_value":"BDMUG3NL","created_at":"2026-07-05T06:23:58Z"}],"graph_snapshots":[{"event_id":"sha256:908e5d345d05beb999119edadac60c4c9efb3f7f6a815e6315b2e7672fdf4ce3","target":"graph","created_at":"2026-07-05T06:23:58Z","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/2301.13464/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"When training deep neural networks, keeping all tensors in high precision (e.g., 32-bit or even 16-bit floats) is often wasteful. However, keeping all tensors in low precision (e.g., 8-bit floats) can lead to unacceptable accuracy loss. Hence, it is important to use a precision assignment -- a mapping from all tensors (arising in training) to precision levels (high or low) -- that keeps most of the tensors in low precision and leads to sufficiently accurate models. We provide a technique that explores this memory-accuracy tradeoff by generating precision assignments for convolutional neural ne","authors_text":"Alex Aiken, Rahul Sharma, Wonyeol Lee","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-01-31T08:01:35Z","title":"Training with Mixed-Precision Floating-Point Assignments"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.13464","kind":"arxiv","version":2},"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":"sha256:19b35d45cc8112701ad91c54bf31aa62f4746f1c557f0c642ee777744ce69e83","target":"record","created_at":"2026-07-05T06:23:58Z","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":"ff0abeb488b1b87a4a3f859de46eda9128589bc5ce59cb485791ff11a690cc0c","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-01-31T08:01:35Z","title_canon_sha256":"0d8fedfde9396f4255fe8edeecc6f62ef6ed4921c9af6cddd0442b3ef3dd529e"},"schema_version":"1.0","source":{"id":"2301.13464","kind":"arxiv","version":2}},"canonical_sha256":"08d9436dab160265e14df04b9e7ccb6e5889876573a75c0e326e5369d4c1f245","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"08d9436dab160265e14df04b9e7ccb6e5889876573a75c0e326e5369d4c1f245","first_computed_at":"2026-07-05T06:23:58.877392Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:23:58.877392Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Sq87j+rtmSgzJ2EOY+oH2C7LD75ivkLsolWoxsZGmiiLPdLThcyqeM2nOf0gKybSaWRONw9jPXE/NE5Oh6zKAA==","signature_status":"signed_v1","signed_at":"2026-07-05T06:23:58.877874Z","signed_message":"canonical_sha256_bytes"},"source_id":"2301.13464","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:19b35d45cc8112701ad91c54bf31aa62f4746f1c557f0c642ee777744ce69e83","sha256:908e5d345d05beb999119edadac60c4c9efb3f7f6a815e6315b2e7672fdf4ce3"],"state_sha256":"0c57856024c5ee431a976d60695f56e3b850c4ee77987c7d864cee3e84499fda"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lJ18KkssRYxVSijj5Zch85dS+R9MOs16IMww9gfL0Q31EVHPsByjPisQu18CcB1qZlAQggHgTFG7hlu3pmQKAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-19T16:10:41.953325Z","bundle_sha256":"7cc56c41a6995bb037c3506606a52fe559e002ef6a67283fec7962bc0b1a9a27"}}