{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:O72FTWY7FJUHPYLPQI6LJOKLK5","short_pith_number":"pith:O72FTWY7","schema_version":"1.0","canonical_sha256":"77f459db1f2a6877e16f823cb4b94b577fbceedb07cfea7a1b3dedf4ace91518","source":{"kind":"arxiv","id":"1908.10396","version":5},"attestation_state":"computed","paper":{"title":"Accelerating Large-Scale Inference with Anisotropic Vector Quantization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"David Simcha, Erik Lindgren, Felix Chern, Philip Sun, Quan Geng, Ruiqi Guo, Sanjiv Kumar","submitted_at":"2019-08-27T18:27:17Z","abstract_excerpt":"Quantization based techniques are the current state-of-the-art for scaling maximum inner product search to massive databases. Traditional approaches to quantization aim to minimize the reconstruction error of the database points. Based on the observation that for a given query, the database points that have the largest inner products are more relevant, we develop a family of anisotropic quantization loss functions. Under natural statistical assumptions, we show that quantization with these loss functions leads to a new variant of vector quantization that more greatly penalizes the parallel com"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"1908.10396","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-08-27T18:27:17Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"3b67d3adaa315a7c1a8f1c7040b496660137f337839817ff1242be2cfcb721eb","abstract_canon_sha256":"cd296adc775496777f86753f55317109b38534be335b253254570b8123135adf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:57:12.705398Z","signature_b64":"D0statEUDO2U+P2+n7RM9CLx1V5aZNn3HMsiGQsqv+cPNRqOM45zyN2FVwcvNIwFqCgTVFyyH3fcyzXtYjntBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"77f459db1f2a6877e16f823cb4b94b577fbceedb07cfea7a1b3dedf4ace91518","last_reissued_at":"2026-07-05T01:57:12.705012Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:57:12.705012Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Accelerating Large-Scale Inference with Anisotropic Vector Quantization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"David Simcha, Erik Lindgren, Felix Chern, Philip Sun, Quan Geng, Ruiqi Guo, Sanjiv Kumar","submitted_at":"2019-08-27T18:27:17Z","abstract_excerpt":"Quantization based techniques are the current state-of-the-art for scaling maximum inner product search to massive databases. Traditional approaches to quantization aim to minimize the reconstruction error of the database points. Based on the observation that for a given query, the database points that have the largest inner products are more relevant, we develop a family of anisotropic quantization loss functions. Under natural statistical assumptions, we show that quantization with these loss functions leads to a new variant of vector quantization that more greatly penalizes the parallel com"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1908.10396","kind":"arxiv","version":5},"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/1908.10396/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"1908.10396","created_at":"2026-07-05T01:57:12.705068+00:00"},{"alias_kind":"arxiv_version","alias_value":"1908.10396v5","created_at":"2026-07-05T01:57:12.705068+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1908.10396","created_at":"2026-07-05T01:57:12.705068+00:00"},{"alias_kind":"pith_short_12","alias_value":"O72FTWY7FJUH","created_at":"2026-07-05T01:57:12.705068+00:00"},{"alias_kind":"pith_short_16","alias_value":"O72FTWY7FJUHPYLP","created_at":"2026-07-05T01:57:12.705068+00:00"},{"alias_kind":"pith_short_8","alias_value":"O72FTWY7","created_at":"2026-07-05T01:57:12.705068+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.12198","citing_title":"LLM-Based User Personas for Recommendations at Scale","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2607.00768","citing_title":"RACORN-1: Adaptive Recall-Preserving Speedup for Low-Selectivity Filtered Vector Search","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2606.31272","citing_title":"The Decomposition Is the Fingerprint: Per-Component Identity for Agent Skills","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16007","citing_title":"Ascend-RaBitQ: Heterogeneous NPU-CPU Acceleration of Billion-Scale Similarity Search with 1-bit Quantization","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2112.04426","citing_title":"Improving language models by retrieving from trillions of tokens","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2603.04545","citing_title":"An LLM-Guided Query-Aware Inference System for GNN Models on Large Knowledge Graphs","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00646","citing_title":"A Replicability Study of XTR","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/O72FTWY7FJUHPYLPQI6LJOKLK5","json":"https://pith.science/pith/O72FTWY7FJUHPYLPQI6LJOKLK5.json","graph_json":"https://pith.science/api/pith-number/O72FTWY7FJUHPYLPQI6LJOKLK5/graph.json","events_json":"https://pith.science/api/pith-number/O72FTWY7FJUHPYLPQI6LJOKLK5/events.json","paper":"https://pith.science/paper/O72FTWY7"},"agent_actions":{"view_html":"https://pith.science/pith/O72FTWY7FJUHPYLPQI6LJOKLK5","download_json":"https://pith.science/pith/O72FTWY7FJUHPYLPQI6LJOKLK5.json","view_paper":"https://pith.science/paper/O72FTWY7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1908.10396&json=true","fetch_graph":"https://pith.science/api/pith-number/O72FTWY7FJUHPYLPQI6LJOKLK5/graph.json","fetch_events":"https://pith.science/api/pith-number/O72FTWY7FJUHPYLPQI6LJOKLK5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/O72FTWY7FJUHPYLPQI6LJOKLK5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/O72FTWY7FJUHPYLPQI6LJOKLK5/action/storage_attestation","attest_author":"https://pith.science/pith/O72FTWY7FJUHPYLPQI6LJOKLK5/action/author_attestation","sign_citation":"https://pith.science/pith/O72FTWY7FJUHPYLPQI6LJOKLK5/action/citation_signature","submit_replication":"https://pith.science/pith/O72FTWY7FJUHPYLPQI6LJOKLK5/action/replication_record"}},"created_at":"2026-07-05T01:57:12.705068+00:00","updated_at":"2026-07-05T01:57:12.705068+00:00"}