{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:I4HFEHFEMQ6VFANGDIOHFI6K7L","short_pith_number":"pith:I4HFEHFE","schema_version":"1.0","canonical_sha256":"470e521ca4643d5281a61a1c72a3cafac1cd274d04f9969efa581531d737f171","source":{"kind":"arxiv","id":"2607.00725","version":1},"attestation_state":"computed","paper":{"title":"What Survives Into Context: A Diagnostic for Budget-Constrained Multi-Hop RAG and When Submodular Evidence Packing Improves It","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Ananto Nayan Bala","submitted_at":"2026-07-01T10:12:15Z","abstract_excerpt":"Retrieval-augmented generation (RAG) under a fixed reader-context budget forces a selection problem: of the evidence retrieved, only a fraction can be shown to the reader. We argue that document recall -- the standard retrieval metric -- is the wrong quantity to optimize in this regime, and we make two contributions. First, as a general contribution, we introduce answer-in-context, a diagnostic that measures whether a gold answer survives as a contiguous span in the packed reader context (not the retrieved set). It predicts answer F1 better than recall (r=0.39-0.55 vs. about 0.31), separates a"},"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":"2607.00725","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-01T10:12:15Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"956af3a25cd0bf7a6cde0d9417a4e431c131161557f52016f622910ed1dde4b5","abstract_canon_sha256":"0ec6d9afbdc10dfce919059739b8495fdec989ad5c2b3142912a7363e516425e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-02T01:17:52.718546Z","signature_b64":"p4Cmp8iY8KiOYq1/xebuoITfmkNqOMQS5i4vRAJ28LvfnISgXQO+X8I/+oADBNrF7ck3/GOkp9bqxrpWgZQ1DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"470e521ca4643d5281a61a1c72a3cafac1cd274d04f9969efa581531d737f171","last_reissued_at":"2026-07-02T01:17:52.718147Z","signature_status":"signed_v1","first_computed_at":"2026-07-02T01:17:52.718147Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"What Survives Into Context: A Diagnostic for Budget-Constrained Multi-Hop RAG and When Submodular Evidence Packing Improves It","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Ananto Nayan Bala","submitted_at":"2026-07-01T10:12:15Z","abstract_excerpt":"Retrieval-augmented generation (RAG) under a fixed reader-context budget forces a selection problem: of the evidence retrieved, only a fraction can be shown to the reader. We argue that document recall -- the standard retrieval metric -- is the wrong quantity to optimize in this regime, and we make two contributions. First, as a general contribution, we introduce answer-in-context, a diagnostic that measures whether a gold answer survives as a contiguous span in the packed reader context (not the retrieved set). It predicts answer F1 better than recall (r=0.39-0.55 vs. about 0.31), separates a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.00725","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/2607.00725/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":"2607.00725","created_at":"2026-07-02T01:17:52.718209+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.00725v1","created_at":"2026-07-02T01:17:52.718209+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.00725","created_at":"2026-07-02T01:17:52.718209+00:00"},{"alias_kind":"pith_short_12","alias_value":"I4HFEHFEMQ6V","created_at":"2026-07-02T01:17:52.718209+00:00"},{"alias_kind":"pith_short_16","alias_value":"I4HFEHFEMQ6VFANG","created_at":"2026-07-02T01:17:52.718209+00:00"},{"alias_kind":"pith_short_8","alias_value":"I4HFEHFE","created_at":"2026-07-02T01:17:52.718209+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/I4HFEHFEMQ6VFANGDIOHFI6K7L","json":"https://pith.science/pith/I4HFEHFEMQ6VFANGDIOHFI6K7L.json","graph_json":"https://pith.science/api/pith-number/I4HFEHFEMQ6VFANGDIOHFI6K7L/graph.json","events_json":"https://pith.science/api/pith-number/I4HFEHFEMQ6VFANGDIOHFI6K7L/events.json","paper":"https://pith.science/paper/I4HFEHFE"},"agent_actions":{"view_html":"https://pith.science/pith/I4HFEHFEMQ6VFANGDIOHFI6K7L","download_json":"https://pith.science/pith/I4HFEHFEMQ6VFANGDIOHFI6K7L.json","view_paper":"https://pith.science/paper/I4HFEHFE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.00725&json=true","fetch_graph":"https://pith.science/api/pith-number/I4HFEHFEMQ6VFANGDIOHFI6K7L/graph.json","fetch_events":"https://pith.science/api/pith-number/I4HFEHFEMQ6VFANGDIOHFI6K7L/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/I4HFEHFEMQ6VFANGDIOHFI6K7L/action/timestamp_anchor","attest_storage":"https://pith.science/pith/I4HFEHFEMQ6VFANGDIOHFI6K7L/action/storage_attestation","attest_author":"https://pith.science/pith/I4HFEHFEMQ6VFANGDIOHFI6K7L/action/author_attestation","sign_citation":"https://pith.science/pith/I4HFEHFEMQ6VFANGDIOHFI6K7L/action/citation_signature","submit_replication":"https://pith.science/pith/I4HFEHFEMQ6VFANGDIOHFI6K7L/action/replication_record"}},"created_at":"2026-07-02T01:17:52.718209+00:00","updated_at":"2026-07-02T01:17:52.718209+00:00"}