{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:QFYNVNFQEGJFAW3GS6PY3S7II4","short_pith_number":"pith:QFYNVNFQ","schema_version":"1.0","canonical_sha256":"8170dab4b02192505b66979f8dcbe8470db1004d2c83b767a29e97ea4e12a35e","source":{"kind":"arxiv","id":"2509.12610","version":2},"attestation_state":"computed","paper":{"title":"ScaleDoc: Scaling LLM-based Predicates over Large Document Collections","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.DB","authors_text":"Hengrui Zhang, Huanchen Zhang, Yihao Liu, Yulong Hui","submitted_at":"2025-09-16T03:18:06Z","abstract_excerpt":"Predicates are foundational components in data analysis systems. However, modern workloads increasingly involve unstructured documents, which demands semantic understanding, beyond traditional value-based predicates. Given enormous documents and ad-hoc queries, while Large Language Models (LLMs) demonstrate powerful zero-shot capabilities, their high inference cost leads to unacceptable overhead. Therefore, we introduce \\textsc{ScaleDoc}, a novel system that addresses this by decoupling predicate execution into an offline representation phase and an optimized online filtering phase. In the off"},"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":"2509.12610","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DB","submitted_at":"2025-09-16T03:18:06Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"8a6fb74f32e2166d7def0b3d0959b37280af42d0c8a0fe4052e1cdd4b09e2a6a","abstract_canon_sha256":"aff35d9191e9f00dd0da27ed387def22481d61110d7450338b10d1d575f54b3b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T01:03:46.589551Z","signature_b64":"oyQ0tGvwjAstx4JwVQrl5PNo6LVDj2hfm9mr49kw5I9xhJrIszrSRwet/Z4+SuHgUX6FL3zJbGb99X5WoB90BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8170dab4b02192505b66979f8dcbe8470db1004d2c83b767a29e97ea4e12a35e","last_reissued_at":"2026-05-22T01:03:46.588587Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T01:03:46.588587Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ScaleDoc: Scaling LLM-based Predicates over Large Document Collections","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.DB","authors_text":"Hengrui Zhang, Huanchen Zhang, Yihao Liu, Yulong Hui","submitted_at":"2025-09-16T03:18:06Z","abstract_excerpt":"Predicates are foundational components in data analysis systems. However, modern workloads increasingly involve unstructured documents, which demands semantic understanding, beyond traditional value-based predicates. Given enormous documents and ad-hoc queries, while Large Language Models (LLMs) demonstrate powerful zero-shot capabilities, their high inference cost leads to unacceptable overhead. Therefore, we introduce \\textsc{ScaleDoc}, a novel system that addresses this by decoupling predicate execution into an offline representation phase and an optimized online filtering phase. In the off"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.12610","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/2509.12610/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":"2509.12610","created_at":"2026-05-22T01:03:46.588738+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.12610v2","created_at":"2026-05-22T01:03:46.588738+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.12610","created_at":"2026-05-22T01:03:46.588738+00:00"},{"alias_kind":"pith_short_12","alias_value":"QFYNVNFQEGJF","created_at":"2026-05-22T01:03:46.588738+00:00"},{"alias_kind":"pith_short_16","alias_value":"QFYNVNFQEGJFAW3G","created_at":"2026-05-22T01:03:46.588738+00:00"},{"alias_kind":"pith_short_8","alias_value":"QFYNVNFQ","created_at":"2026-05-22T01:03:46.588738+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2604.09944","citing_title":"PLOP: Cost-Based Placement of Semantic Operators in Hybrid Query Plans","ref_index":34,"is_internal_anchor":true},{"citing_arxiv_id":"2604.21072","citing_title":"Distributed Generative Inference of LLM at Internet Scales with Multi-Dimensional Communication Optimization","ref_index":48,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QFYNVNFQEGJFAW3GS6PY3S7II4","json":"https://pith.science/pith/QFYNVNFQEGJFAW3GS6PY3S7II4.json","graph_json":"https://pith.science/api/pith-number/QFYNVNFQEGJFAW3GS6PY3S7II4/graph.json","events_json":"https://pith.science/api/pith-number/QFYNVNFQEGJFAW3GS6PY3S7II4/events.json","paper":"https://pith.science/paper/QFYNVNFQ"},"agent_actions":{"view_html":"https://pith.science/pith/QFYNVNFQEGJFAW3GS6PY3S7II4","download_json":"https://pith.science/pith/QFYNVNFQEGJFAW3GS6PY3S7II4.json","view_paper":"https://pith.science/paper/QFYNVNFQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.12610&json=true","fetch_graph":"https://pith.science/api/pith-number/QFYNVNFQEGJFAW3GS6PY3S7II4/graph.json","fetch_events":"https://pith.science/api/pith-number/QFYNVNFQEGJFAW3GS6PY3S7II4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QFYNVNFQEGJFAW3GS6PY3S7II4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QFYNVNFQEGJFAW3GS6PY3S7II4/action/storage_attestation","attest_author":"https://pith.science/pith/QFYNVNFQEGJFAW3GS6PY3S7II4/action/author_attestation","sign_citation":"https://pith.science/pith/QFYNVNFQEGJFAW3GS6PY3S7II4/action/citation_signature","submit_replication":"https://pith.science/pith/QFYNVNFQEGJFAW3GS6PY3S7II4/action/replication_record"}},"created_at":"2026-05-22T01:03:46.588738+00:00","updated_at":"2026-05-22T01:03:46.588738+00:00"}