{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:RCG47T2T5CASC55KVC74OUHLMR","short_pith_number":"pith:RCG47T2T","schema_version":"1.0","canonical_sha256":"888dcfcf53e8812177aaa8bfc750eb6464eb74ef8c2474deefbf41840bbb7c10","source":{"kind":"arxiv","id":"2607.05764","version":1},"attestation_state":"computed","paper":{"title":"Inject or Navigate? Token-Efficient Retrieval for LLM Analysis of Transactional Legal Documents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Mahmoud Hany, Mahmoud Said, Mourad ElSheraey, Peter Naoum","submitted_at":"2026-07-07T02:42:06Z","abstract_excerpt":"Answering questions over a set of transactional legal documents is most simply done by injecting the whole corpus into the LLM's context window on every query. That baseline maximises retrieval recall, but its token footprint scales with the corpus rather than the question, and long-context degradation scales with it. We report what it took to replace full-corpus injection in a legal-document analysis system, comparing it against two structured retrieval modes over our proprietary structure-aware chunking: embedding retrieval (NAVEMBED) and LLM navigation over a compact structured index (NAVIN"},"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":"2607.05764","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-07T02:42:06Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"62975739055bd774d8cb474530846b7dde57641d17bc336d1b34b476681823d0","abstract_canon_sha256":"6326b6fc226dcd313123c64904e4a851c2e41e7d4d431ceff42f1002806f422d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-08T01:18:44.185164Z","signature_b64":"ITcZLqqfi3hLScmgYafTge6nHKlzrGP6IjDx/mpEO2wReBpGa7VSBo2Fxy8qetDdi9NxGPf3VrELE50oq35nBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"888dcfcf53e8812177aaa8bfc750eb6464eb74ef8c2474deefbf41840bbb7c10","last_reissued_at":"2026-07-08T01:18:44.184510Z","signature_status":"signed_v1","first_computed_at":"2026-07-08T01:18:44.184510Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Inject or Navigate? Token-Efficient Retrieval for LLM Analysis of Transactional Legal Documents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Mahmoud Hany, Mahmoud Said, Mourad ElSheraey, Peter Naoum","submitted_at":"2026-07-07T02:42:06Z","abstract_excerpt":"Answering questions over a set of transactional legal documents is most simply done by injecting the whole corpus into the LLM's context window on every query. That baseline maximises retrieval recall, but its token footprint scales with the corpus rather than the question, and long-context degradation scales with it. We report what it took to replace full-corpus injection in a legal-document analysis system, comparing it against two structured retrieval modes over our proprietary structure-aware chunking: embedding retrieval (NAVEMBED) and LLM navigation over a compact structured index (NAVIN"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.05764","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/2607.05764/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":"2607.05764","created_at":"2026-07-08T01:18:44.184611+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.05764v1","created_at":"2026-07-08T01:18:44.184611+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.05764","created_at":"2026-07-08T01:18:44.184611+00:00"},{"alias_kind":"pith_short_12","alias_value":"RCG47T2T5CAS","created_at":"2026-07-08T01:18:44.184611+00:00"},{"alias_kind":"pith_short_16","alias_value":"RCG47T2T5CASC55K","created_at":"2026-07-08T01:18:44.184611+00:00"},{"alias_kind":"pith_short_8","alias_value":"RCG47T2T","created_at":"2026-07-08T01:18:44.184611+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RCG47T2T5CASC55KVC74OUHLMR","json":"https://pith.science/pith/RCG47T2T5CASC55KVC74OUHLMR.json","graph_json":"https://pith.science/api/pith-number/RCG47T2T5CASC55KVC74OUHLMR/graph.json","events_json":"https://pith.science/api/pith-number/RCG47T2T5CASC55KVC74OUHLMR/events.json","paper":"https://pith.science/paper/RCG47T2T"},"agent_actions":{"view_html":"https://pith.science/pith/RCG47T2T5CASC55KVC74OUHLMR","download_json":"https://pith.science/pith/RCG47T2T5CASC55KVC74OUHLMR.json","view_paper":"https://pith.science/paper/RCG47T2T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.05764&json=true","fetch_graph":"https://pith.science/api/pith-number/RCG47T2T5CASC55KVC74OUHLMR/graph.json","fetch_events":"https://pith.science/api/pith-number/RCG47T2T5CASC55KVC74OUHLMR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RCG47T2T5CASC55KVC74OUHLMR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RCG47T2T5CASC55KVC74OUHLMR/action/storage_attestation","attest_author":"https://pith.science/pith/RCG47T2T5CASC55KVC74OUHLMR/action/author_attestation","sign_citation":"https://pith.science/pith/RCG47T2T5CASC55KVC74OUHLMR/action/citation_signature","submit_replication":"https://pith.science/pith/RCG47T2T5CASC55KVC74OUHLMR/action/replication_record"}},"created_at":"2026-07-08T01:18:44.184611+00:00","updated_at":"2026-07-08T01:18:44.184611+00:00"}