{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:TS4ZVW3TRZQETOKQZ46XHA7N3T","short_pith_number":"pith:TS4ZVW3T","canonical_record":{"source":{"id":"2507.01676","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-07-02T13:00:39Z","cross_cats_sorted":["cs.AI","cs.AR","cs.IR"],"title_canon_sha256":"6acffeb94a213aaa168947088d7748fb3ca76f3d00b41cac5ea02e304cb4ce7c","abstract_canon_sha256":"4900851c0ee67e0b4ad05cf81457016cc3c9c514e1a678ffa4135b3954beaa8a"},"schema_version":"1.0"},"canonical_sha256":"9cb99adb738e6049b950cf3d7383eddcd30fafbb5e198c68a9d987a5abc37c40","source":{"kind":"arxiv","id":"2507.01676","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.01676","created_at":"2026-07-05T11:30:53Z"},{"alias_kind":"arxiv_version","alias_value":"2507.01676v1","created_at":"2026-07-05T11:30:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.01676","created_at":"2026-07-05T11:30:53Z"},{"alias_kind":"pith_short_12","alias_value":"TS4ZVW3TRZQE","created_at":"2026-07-05T11:30:53Z"},{"alias_kind":"pith_short_16","alias_value":"TS4ZVW3TRZQETOKQ","created_at":"2026-07-05T11:30:53Z"},{"alias_kind":"pith_short_8","alias_value":"TS4ZVW3T","created_at":"2026-07-05T11:30:53Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:TS4ZVW3TRZQETOKQZ46XHA7N3T","target":"record","payload":{"canonical_record":{"source":{"id":"2507.01676","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-07-02T13:00:39Z","cross_cats_sorted":["cs.AI","cs.AR","cs.IR"],"title_canon_sha256":"6acffeb94a213aaa168947088d7748fb3ca76f3d00b41cac5ea02e304cb4ce7c","abstract_canon_sha256":"4900851c0ee67e0b4ad05cf81457016cc3c9c514e1a678ffa4135b3954beaa8a"},"schema_version":"1.0"},"canonical_sha256":"9cb99adb738e6049b950cf3d7383eddcd30fafbb5e198c68a9d987a5abc37c40","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:30:53.079907Z","signature_b64":"wgenVi7AZgqny3nlsrXnFMTtUOtibreHpe2ee81vaZuduDjw4vzCqB8RxlSfmkBtwVilgxqQ6pvy13GwY6LDAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9cb99adb738e6049b950cf3d7383eddcd30fafbb5e198c68a9d987a5abc37c40","last_reissued_at":"2026-07-05T11:30:53.079398Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:30:53.079398Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.01676","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:30:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zc5CNgetUfFfFTZ1o7NL5FHd63p+V1h4DvpX1DaP7cqcyCVl6wS2Eq73go+XZFTJRCGFxCDA65m6odSkUpQ9CQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T12:11:53.966424Z"},"content_sha256":"518460ab4bb7b290f188c7f1f40b58a689a39727e98239790124ab3d5dda6513","schema_version":"1.0","event_id":"sha256:518460ab4bb7b290f188c7f1f40b58a689a39727e98239790124ab3d5dda6513"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:TS4ZVW3TRZQETOKQZ46XHA7N3T","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Deep Recommender Models Inference: Automatic Asymmetric Data Flow Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.AR","cs.IR"],"primary_cat":"cs.DC","authors_text":"Daniele Jahier Pagliari, Giuseppe Ruggeri, Lukas Cavigelli, Renzo Andri","submitted_at":"2025-07-02T13:00:39Z","abstract_excerpt":"Deep Recommender Models (DLRMs) inference is a fundamental AI workload accounting for more than 79% of the total AI workload in Meta's data centers. DLRMs' performance bottleneck is found in the embedding layers, which perform many random memory accesses to retrieve small embedding vectors from tables of various sizes. We propose the design of tailored data flows to speedup embedding look-ups. Namely, we propose four strategies to look up an embedding table effectively on one core, and a framework to automatically map the tables asymmetrically to the multiple cores of a SoC. We assess the effe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.01676","kind":"arxiv","version":1},"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/2507.01676/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":"2026-07-05T11:30:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ht0BQF6T59wwO4d0pUu43h2JAs2oGaWqdicPOd7wkQA5b6hqYQxKoDf86bLeLdH23VBigEunfJnLLhSsSKOJBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T12:11:53.967061Z"},"content_sha256":"364c14e8a8c6ed210d842572af210ac6a1984b3219c6d8ceb81669f295158ff8","schema_version":"1.0","event_id":"sha256:364c14e8a8c6ed210d842572af210ac6a1984b3219c6d8ceb81669f295158ff8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TS4ZVW3TRZQETOKQZ46XHA7N3T/bundle.json","state_url":"https://pith.science/pith/TS4ZVW3TRZQETOKQZ46XHA7N3T/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TS4ZVW3TRZQETOKQZ46XHA7N3T/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":"2026-08-10T12:11:53Z","links":{"resolver":"https://pith.science/pith/TS4ZVW3TRZQETOKQZ46XHA7N3T","bundle":"https://pith.science/pith/TS4ZVW3TRZQETOKQZ46XHA7N3T/bundle.json","state":"https://pith.science/pith/TS4ZVW3TRZQETOKQZ46XHA7N3T/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TS4ZVW3TRZQETOKQZ46XHA7N3T/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:TS4ZVW3TRZQETOKQZ46XHA7N3T","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":"4900851c0ee67e0b4ad05cf81457016cc3c9c514e1a678ffa4135b3954beaa8a","cross_cats_sorted":["cs.AI","cs.AR","cs.IR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-07-02T13:00:39Z","title_canon_sha256":"6acffeb94a213aaa168947088d7748fb3ca76f3d00b41cac5ea02e304cb4ce7c"},"schema_version":"1.0","source":{"id":"2507.01676","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.01676","created_at":"2026-07-05T11:30:53Z"},{"alias_kind":"arxiv_version","alias_value":"2507.01676v1","created_at":"2026-07-05T11:30:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.01676","created_at":"2026-07-05T11:30:53Z"},{"alias_kind":"pith_short_12","alias_value":"TS4ZVW3TRZQE","created_at":"2026-07-05T11:30:53Z"},{"alias_kind":"pith_short_16","alias_value":"TS4ZVW3TRZQETOKQ","created_at":"2026-07-05T11:30:53Z"},{"alias_kind":"pith_short_8","alias_value":"TS4ZVW3T","created_at":"2026-07-05T11:30:53Z"}],"graph_snapshots":[{"event_id":"sha256:364c14e8a8c6ed210d842572af210ac6a1984b3219c6d8ceb81669f295158ff8","target":"graph","created_at":"2026-07-05T11:30:53Z","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/2507.01676/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Deep Recommender Models (DLRMs) inference is a fundamental AI workload accounting for more than 79% of the total AI workload in Meta's data centers. DLRMs' performance bottleneck is found in the embedding layers, which perform many random memory accesses to retrieve small embedding vectors from tables of various sizes. We propose the design of tailored data flows to speedup embedding look-ups. Namely, we propose four strategies to look up an embedding table effectively on one core, and a framework to automatically map the tables asymmetrically to the multiple cores of a SoC. We assess the effe","authors_text":"Daniele Jahier Pagliari, Giuseppe Ruggeri, Lukas Cavigelli, Renzo Andri","cross_cats":["cs.AI","cs.AR","cs.IR"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-07-02T13:00:39Z","title":"Deep Recommender Models Inference: Automatic Asymmetric Data Flow Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.01676","kind":"arxiv","version":1},"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":"sha256:518460ab4bb7b290f188c7f1f40b58a689a39727e98239790124ab3d5dda6513","target":"record","created_at":"2026-07-05T11:30:53Z","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":"4900851c0ee67e0b4ad05cf81457016cc3c9c514e1a678ffa4135b3954beaa8a","cross_cats_sorted":["cs.AI","cs.AR","cs.IR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-07-02T13:00:39Z","title_canon_sha256":"6acffeb94a213aaa168947088d7748fb3ca76f3d00b41cac5ea02e304cb4ce7c"},"schema_version":"1.0","source":{"id":"2507.01676","kind":"arxiv","version":1}},"canonical_sha256":"9cb99adb738e6049b950cf3d7383eddcd30fafbb5e198c68a9d987a5abc37c40","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9cb99adb738e6049b950cf3d7383eddcd30fafbb5e198c68a9d987a5abc37c40","first_computed_at":"2026-07-05T11:30:53.079398Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:30:53.079398Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"wgenVi7AZgqny3nlsrXnFMTtUOtibreHpe2ee81vaZuduDjw4vzCqB8RxlSfmkBtwVilgxqQ6pvy13GwY6LDAA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:30:53.079907Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.01676","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:518460ab4bb7b290f188c7f1f40b58a689a39727e98239790124ab3d5dda6513","sha256:364c14e8a8c6ed210d842572af210ac6a1984b3219c6d8ceb81669f295158ff8"],"state_sha256":"578682a560bb1b2580f406b8713cf7557f1e633a2b06333fa56ebddb038014b5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"I5cDZGr2GADY2VugoEdIDTWeUY3jNEpZISyHz/BlBVfYkuZtMct54WaKitDINi6doRg8WVooZVZ54FfvSiYeDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T12:11:53.973744Z","bundle_sha256":"164d1c344987f3434b385b7d870db3d004e83c0abb35bb761102303357f0956d"}}