{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:KPY4OR3QIOJTRP2G6OKHOSOQXP","short_pith_number":"pith:KPY4OR3Q","canonical_record":{"source":{"id":"1909.02107","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-09-04T20:47:08Z","cross_cats_sorted":["cs.IR","stat.ML"],"title_canon_sha256":"011f1eb1534de0f6254f8042025139ddda8d156318912db270b9c37b6770e658","abstract_canon_sha256":"4be7f48ab23f49963a2b5bd5ff45827b194c08ee3791fea51f139e7c62f0d5ef"},"schema_version":"1.0"},"canonical_sha256":"53f1c74770439338bf46f3947749d0bbc3ac01d7d9d0807a1c419558a4c4e3a8","source":{"kind":"arxiv","id":"1909.02107","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1909.02107","created_at":"2026-07-05T01:13:54Z"},{"alias_kind":"arxiv_version","alias_value":"1909.02107v2","created_at":"2026-07-05T01:13:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1909.02107","created_at":"2026-07-05T01:13:54Z"},{"alias_kind":"pith_short_12","alias_value":"KPY4OR3QIOJT","created_at":"2026-07-05T01:13:54Z"},{"alias_kind":"pith_short_16","alias_value":"KPY4OR3QIOJTRP2G","created_at":"2026-07-05T01:13:54Z"},{"alias_kind":"pith_short_8","alias_value":"KPY4OR3Q","created_at":"2026-07-05T01:13:54Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:KPY4OR3QIOJTRP2G6OKHOSOQXP","target":"record","payload":{"canonical_record":{"source":{"id":"1909.02107","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-09-04T20:47:08Z","cross_cats_sorted":["cs.IR","stat.ML"],"title_canon_sha256":"011f1eb1534de0f6254f8042025139ddda8d156318912db270b9c37b6770e658","abstract_canon_sha256":"4be7f48ab23f49963a2b5bd5ff45827b194c08ee3791fea51f139e7c62f0d5ef"},"schema_version":"1.0"},"canonical_sha256":"53f1c74770439338bf46f3947749d0bbc3ac01d7d9d0807a1c419558a4c4e3a8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:13:54.542750Z","signature_b64":"U9oK9mMEv9KhxSa48bXeA342cK+Kka5BXExlQ7cPbsxxWomNKqoLh5S9XF+W4vsvcxYY4KAT1ki1xdyeusV1AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"53f1c74770439338bf46f3947749d0bbc3ac01d7d9d0807a1c419558a4c4e3a8","last_reissued_at":"2026-07-05T01:13:54.542255Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:13:54.542255Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1909.02107","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":"2026-07-05T01:13:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LyYNGyxhksvX7i7KeGukUVb5RdWs/fqn917YgjR8G1QLqcVndMQNPZkRpWw6PtchZ5oy8K3ibpe2ZpddvLAEBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T19:09:59.982364Z"},"content_sha256":"b36704a95a51f60d8c5a2cc8bd1344a66de3dc40afa18e56a77433e7357bb451","schema_version":"1.0","event_id":"sha256:b36704a95a51f60d8c5a2cc8bd1344a66de3dc40afa18e56a77433e7357bb451"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:KPY4OR3QIOJTRP2G6OKHOSOQXP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Compositional Embeddings Using Complementary Partitions for Memory-Efficient Recommendation Systems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.IR","stat.ML"],"primary_cat":"cs.LG","authors_text":"Dheevatsa Mudigere, Hao-Jun Michael Shi, Jiyan Yang, Maxim Naumov","submitted_at":"2019-09-04T20:47:08Z","abstract_excerpt":"Modern deep learning-based recommendation systems exploit hundreds to thousands of different categorical features, each with millions of different categories ranging from clicks to posts. To respect the natural diversity within the categorical data, embeddings map each category to a unique dense representation within an embedded space. Since each categorical feature could take on as many as tens of millions of different possible categories, the embedding tables form the primary memory bottleneck during both training and inference. We propose a novel approach for reducing the embedding size in "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1909.02107","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/1909.02107/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":"2026-07-05T01:13:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cppLQ/po4KK3FQ7Yx5KmjArhF92bFRk0wdXwTdiqhJMfwLJPfH39ZGv3HAEXtLr+SSW1N3+KqJm1y9HVqHwEDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T19:09:59.982905Z"},"content_sha256":"9ae590ef1dbc3a388703af38b41ceb9c95b15584f6bf58e802ce09b1b15f2855","schema_version":"1.0","event_id":"sha256:9ae590ef1dbc3a388703af38b41ceb9c95b15584f6bf58e802ce09b1b15f2855"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KPY4OR3QIOJTRP2G6OKHOSOQXP/bundle.json","state_url":"https://pith.science/pith/KPY4OR3QIOJTRP2G6OKHOSOQXP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KPY4OR3QIOJTRP2G6OKHOSOQXP/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":"2026-08-14T19:09:59Z","links":{"resolver":"https://pith.science/pith/KPY4OR3QIOJTRP2G6OKHOSOQXP","bundle":"https://pith.science/pith/KPY4OR3QIOJTRP2G6OKHOSOQXP/bundle.json","state":"https://pith.science/pith/KPY4OR3QIOJTRP2G6OKHOSOQXP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KPY4OR3QIOJTRP2G6OKHOSOQXP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:KPY4OR3QIOJTRP2G6OKHOSOQXP","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":"4be7f48ab23f49963a2b5bd5ff45827b194c08ee3791fea51f139e7c62f0d5ef","cross_cats_sorted":["cs.IR","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-09-04T20:47:08Z","title_canon_sha256":"011f1eb1534de0f6254f8042025139ddda8d156318912db270b9c37b6770e658"},"schema_version":"1.0","source":{"id":"1909.02107","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1909.02107","created_at":"2026-07-05T01:13:54Z"},{"alias_kind":"arxiv_version","alias_value":"1909.02107v2","created_at":"2026-07-05T01:13:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1909.02107","created_at":"2026-07-05T01:13:54Z"},{"alias_kind":"pith_short_12","alias_value":"KPY4OR3QIOJT","created_at":"2026-07-05T01:13:54Z"},{"alias_kind":"pith_short_16","alias_value":"KPY4OR3QIOJTRP2G","created_at":"2026-07-05T01:13:54Z"},{"alias_kind":"pith_short_8","alias_value":"KPY4OR3Q","created_at":"2026-07-05T01:13:54Z"}],"graph_snapshots":[{"event_id":"sha256:9ae590ef1dbc3a388703af38b41ceb9c95b15584f6bf58e802ce09b1b15f2855","target":"graph","created_at":"2026-07-05T01:13:54Z","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/1909.02107/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Modern deep learning-based recommendation systems exploit hundreds to thousands of different categorical features, each with millions of different categories ranging from clicks to posts. To respect the natural diversity within the categorical data, embeddings map each category to a unique dense representation within an embedded space. Since each categorical feature could take on as many as tens of millions of different possible categories, the embedding tables form the primary memory bottleneck during both training and inference. We propose a novel approach for reducing the embedding size in ","authors_text":"Dheevatsa Mudigere, Hao-Jun Michael Shi, Jiyan Yang, Maxim Naumov","cross_cats":["cs.IR","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-09-04T20:47:08Z","title":"Compositional Embeddings Using Complementary Partitions for Memory-Efficient Recommendation Systems"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1909.02107","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:b36704a95a51f60d8c5a2cc8bd1344a66de3dc40afa18e56a77433e7357bb451","target":"record","created_at":"2026-07-05T01:13:54Z","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":"4be7f48ab23f49963a2b5bd5ff45827b194c08ee3791fea51f139e7c62f0d5ef","cross_cats_sorted":["cs.IR","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-09-04T20:47:08Z","title_canon_sha256":"011f1eb1534de0f6254f8042025139ddda8d156318912db270b9c37b6770e658"},"schema_version":"1.0","source":{"id":"1909.02107","kind":"arxiv","version":2}},"canonical_sha256":"53f1c74770439338bf46f3947749d0bbc3ac01d7d9d0807a1c419558a4c4e3a8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"53f1c74770439338bf46f3947749d0bbc3ac01d7d9d0807a1c419558a4c4e3a8","first_computed_at":"2026-07-05T01:13:54.542255Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T01:13:54.542255Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"U9oK9mMEv9KhxSa48bXeA342cK+Kka5BXExlQ7cPbsxxWomNKqoLh5S9XF+W4vsvcxYY4KAT1ki1xdyeusV1AQ==","signature_status":"signed_v1","signed_at":"2026-07-05T01:13:54.542750Z","signed_message":"canonical_sha256_bytes"},"source_id":"1909.02107","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b36704a95a51f60d8c5a2cc8bd1344a66de3dc40afa18e56a77433e7357bb451","sha256:9ae590ef1dbc3a388703af38b41ceb9c95b15584f6bf58e802ce09b1b15f2855"],"state_sha256":"bb64445b7a5ebb04cdb515fe9dd5300a2175e921d65c77a248b2e0cdbc5f3ffb"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+VGhpfWLiaZw1A75Q437S4moZSyjlxG9FeKZ8lpnqnzjCSnvyFyUkBysLTPhYIvOI1Tc7HgB6SoYdE3HVA2yCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T19:09:59.987277Z","bundle_sha256":"324e894274e9cceb4b348811a5df24a4a44eedb31b97ff52e13e0320a11bf97b"}}