{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2016:ITXFMALYUNRQFFQTYDIN3HWSAL","short_pith_number":"pith:ITXFMALY","canonical_record":{"source":{"id":"1611.05402","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-11-16T18:45:09Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"4bb3bed73cc52a9fcf8f5b0354a81a642e760afefb25c82d4239d7817d94ca75","abstract_canon_sha256":"e542191ef32f2e014907c96fed7e026a3b878653813dfc04903c1625d4fda602"},"schema_version":"1.0"},"canonical_sha256":"44ee560178a363029613c0d0dd9ed202edebff537b05b4754f25749615eec753","source":{"kind":"arxiv","id":"1611.05402","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1611.05402","created_at":"2026-05-18T00:42:12Z"},{"alias_kind":"arxiv_version","alias_value":"1611.05402v3","created_at":"2026-05-18T00:42:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1611.05402","created_at":"2026-05-18T00:42:12Z"},{"alias_kind":"pith_short_12","alias_value":"ITXFMALYUNRQ","created_at":"2026-05-18T12:30:22Z"},{"alias_kind":"pith_short_16","alias_value":"ITXFMALYUNRQFFQT","created_at":"2026-05-18T12:30:22Z"},{"alias_kind":"pith_short_8","alias_value":"ITXFMALY","created_at":"2026-05-18T12:30:22Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2016:ITXFMALYUNRQFFQTYDIN3HWSAL","target":"record","payload":{"canonical_record":{"source":{"id":"1611.05402","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-11-16T18:45:09Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"4bb3bed73cc52a9fcf8f5b0354a81a642e760afefb25c82d4239d7817d94ca75","abstract_canon_sha256":"e542191ef32f2e014907c96fed7e026a3b878653813dfc04903c1625d4fda602"},"schema_version":"1.0"},"canonical_sha256":"44ee560178a363029613c0d0dd9ed202edebff537b05b4754f25749615eec753","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:42:12.669688Z","signature_b64":"EI1Z2k1xcTDJgEcRProsYB6Q0uoI6nSZS5vmRDSEzQQgeggJk5M4TXfwnzb+use6ZTEI2lggV3dcbXPeIeBFAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"44ee560178a363029613c0d0dd9ed202edebff537b05b4754f25749615eec753","last_reissued_at":"2026-05-18T00:42:12.669282Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:42:12.669282Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1611.05402","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":"2026-05-18T00:42:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WeBzx2BBg6iB4X+XzWxlkhH1syMONKeWED41Rs7CbltzO52aWvm5jAUe2bcu7r/uZldtosz9LduQE4GHzI8YDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T11:07:56.756062Z"},"content_sha256":"a9cccf343840123978f52fdaa4a2554092584af955739a78c50daa8e208015f3","schema_version":"1.0","event_id":"sha256:a9cccf343840123978f52fdaa4a2554092584af955739a78c50daa8e208015f3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2016:ITXFMALYUNRQFFQTYDIN3HWSAL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"The ZipML Framework for Training Models with End-to-End Low Precision: The Cans, the Cannots, and a Little Bit of Deep Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Ce Zhang, Dan Alistarh, Hantian Zhang, Jerry Li, Ji Liu, Kaan Kara","submitted_at":"2016-11-16T18:45:09Z","abstract_excerpt":"Recently there has been significant interest in training machine-learning models at low precision: by reducing precision, one can reduce computation and communication by one order of magnitude. We examine training at reduced precision, both from a theoretical and practical perspective, and ask: is it possible to train models at end-to-end low precision with provable guarantees? Can this lead to consistent order-of-magnitude speedups? We present a framework called ZipML to answer these questions. For linear models, the answer is yes. We develop a simple framework based on one simple but novel s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1611.05402","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":""},"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":"2026-05-18T00:42:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"z6vBWXZjtAvkhHnXZOBAbMnxX8xsOH/VFtEhXqawsUkc/rhyAaiWR4M8w1mCso2uUiz4FiWvSlE06FjI8VbDDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T11:07:56.756715Z"},"content_sha256":"4b8ff21b844fc15521cf20b47c1f644914ee6d28b0b716fabeb789bcb9bace7d","schema_version":"1.0","event_id":"sha256:4b8ff21b844fc15521cf20b47c1f644914ee6d28b0b716fabeb789bcb9bace7d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ITXFMALYUNRQFFQTYDIN3HWSAL/bundle.json","state_url":"https://pith.science/pith/ITXFMALYUNRQFFQTYDIN3HWSAL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ITXFMALYUNRQFFQTYDIN3HWSAL/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":"2026-05-26T11:07:56Z","links":{"resolver":"https://pith.science/pith/ITXFMALYUNRQFFQTYDIN3HWSAL","bundle":"https://pith.science/pith/ITXFMALYUNRQFFQTYDIN3HWSAL/bundle.json","state":"https://pith.science/pith/ITXFMALYUNRQFFQTYDIN3HWSAL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ITXFMALYUNRQFFQTYDIN3HWSAL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2016:ITXFMALYUNRQFFQTYDIN3HWSAL","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":"e542191ef32f2e014907c96fed7e026a3b878653813dfc04903c1625d4fda602","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-11-16T18:45:09Z","title_canon_sha256":"4bb3bed73cc52a9fcf8f5b0354a81a642e760afefb25c82d4239d7817d94ca75"},"schema_version":"1.0","source":{"id":"1611.05402","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1611.05402","created_at":"2026-05-18T00:42:12Z"},{"alias_kind":"arxiv_version","alias_value":"1611.05402v3","created_at":"2026-05-18T00:42:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1611.05402","created_at":"2026-05-18T00:42:12Z"},{"alias_kind":"pith_short_12","alias_value":"ITXFMALYUNRQ","created_at":"2026-05-18T12:30:22Z"},{"alias_kind":"pith_short_16","alias_value":"ITXFMALYUNRQFFQT","created_at":"2026-05-18T12:30:22Z"},{"alias_kind":"pith_short_8","alias_value":"ITXFMALY","created_at":"2026-05-18T12:30:22Z"}],"graph_snapshots":[{"event_id":"sha256:4b8ff21b844fc15521cf20b47c1f644914ee6d28b0b716fabeb789bcb9bace7d","target":"graph","created_at":"2026-05-18T00:42:12Z","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"},"paper":{"abstract_excerpt":"Recently there has been significant interest in training machine-learning models at low precision: by reducing precision, one can reduce computation and communication by one order of magnitude. We examine training at reduced precision, both from a theoretical and practical perspective, and ask: is it possible to train models at end-to-end low precision with provable guarantees? Can this lead to consistent order-of-magnitude speedups? We present a framework called ZipML to answer these questions. For linear models, the answer is yes. We develop a simple framework based on one simple but novel s","authors_text":"Ce Zhang, Dan Alistarh, Hantian Zhang, Jerry Li, Ji Liu, Kaan Kara","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-11-16T18:45:09Z","title":"The ZipML Framework for Training Models with End-to-End Low Precision: The Cans, the Cannots, and a Little Bit of Deep Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1611.05402","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:a9cccf343840123978f52fdaa4a2554092584af955739a78c50daa8e208015f3","target":"record","created_at":"2026-05-18T00:42:12Z","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":"e542191ef32f2e014907c96fed7e026a3b878653813dfc04903c1625d4fda602","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-11-16T18:45:09Z","title_canon_sha256":"4bb3bed73cc52a9fcf8f5b0354a81a642e760afefb25c82d4239d7817d94ca75"},"schema_version":"1.0","source":{"id":"1611.05402","kind":"arxiv","version":3}},"canonical_sha256":"44ee560178a363029613c0d0dd9ed202edebff537b05b4754f25749615eec753","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"44ee560178a363029613c0d0dd9ed202edebff537b05b4754f25749615eec753","first_computed_at":"2026-05-18T00:42:12.669282Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:42:12.669282Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"EI1Z2k1xcTDJgEcRProsYB6Q0uoI6nSZS5vmRDSEzQQgeggJk5M4TXfwnzb+use6ZTEI2lggV3dcbXPeIeBFAA==","signature_status":"signed_v1","signed_at":"2026-05-18T00:42:12.669688Z","signed_message":"canonical_sha256_bytes"},"source_id":"1611.05402","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a9cccf343840123978f52fdaa4a2554092584af955739a78c50daa8e208015f3","sha256:4b8ff21b844fc15521cf20b47c1f644914ee6d28b0b716fabeb789bcb9bace7d"],"state_sha256":"cf043f2516eed0ece5b35689f8ae1f7ca9906e621e2496ed7c1c6d5cff0ebd13"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xHwwW2EzPZ70ym5DiwOWo3cbgUR71iAQgoWQZtWAzurDxsEqvLHsSzNCp1XBOVELM+09OTis1M9bYQInFXX/Bg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-26T11:07:56.760193Z","bundle_sha256":"645432d86e68961592c2ea766598c774528faff7e85cf263a5fffccf8b573dad"}}