{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:PWIZRP5EM5AFJ7CFQ4XY77VFDL","short_pith_number":"pith:PWIZRP5E","schema_version":"1.0","canonical_sha256":"7d9198bfa4674054fc45872f8ffea51af4db1f04037d0a75cd146b6ef6448082","source":{"kind":"arxiv","id":"2605.22219","version":1},"attestation_state":"computed","paper":{"title":"SGR-Bench: Benchmarking Search Agents on State-Gated Retrieval","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Haiyang Shen, Mugeng Liu, Ningyuan Li, Sixiong Xie, Yudong Han, Yun Ma, Zhuofan Shi","submitted_at":"2026-05-21T09:22:48Z","abstract_excerpt":"Recent advances in large language models and tool-using agents have expanded the range of benchmarked web tasks. Yet an important class of specialized retrieval tasks remains undercharacterized. On many specialized data-retrieval websites, answer-bearing evidence becomes accessible only after establishing the correct site-specific retrieval state through filters, views, hierarchies, or scopes. We term this capability state-gated retrieval (SGR). We introduce SGR-Bench, a benchmark for this setting containing 100 expert-curated tasks spanning six source families and 12 public data ecosystems. E"},"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":"2605.22219","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-21T09:22:48Z","cross_cats_sorted":[],"title_canon_sha256":"58392aa1629dfa0573a47914930cf4ff2ddbed8862a35a4d683454f87987e895","abstract_canon_sha256":"82f1cf8eda1cfacfff5a53c4dfc3d9f02f3d9e0b86a32672120943e9ac87be9c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T01:04:32.764910Z","signature_b64":"Qr9XJFz/+Riy5SPR98yg7luIs+VFqL7iGJOkPaxqozdi/uN6v09h9kO39ugRjMODW8vFn4ynGzP7LlRK+9goAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7d9198bfa4674054fc45872f8ffea51af4db1f04037d0a75cd146b6ef6448082","last_reissued_at":"2026-05-22T01:04:32.764247Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T01:04:32.764247Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SGR-Bench: Benchmarking Search Agents on State-Gated Retrieval","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Haiyang Shen, Mugeng Liu, Ningyuan Li, Sixiong Xie, Yudong Han, Yun Ma, Zhuofan Shi","submitted_at":"2026-05-21T09:22:48Z","abstract_excerpt":"Recent advances in large language models and tool-using agents have expanded the range of benchmarked web tasks. Yet an important class of specialized retrieval tasks remains undercharacterized. On many specialized data-retrieval websites, answer-bearing evidence becomes accessible only after establishing the correct site-specific retrieval state through filters, views, hierarchies, or scopes. We term this capability state-gated retrieval (SGR). We introduce SGR-Bench, a benchmark for this setting containing 100 expert-curated tasks spanning six source families and 12 public data ecosystems. E"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.22219","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/2605.22219/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":"2605.22219","created_at":"2026-05-22T01:04:32.764358+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.22219v1","created_at":"2026-05-22T01:04:32.764358+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.22219","created_at":"2026-05-22T01:04:32.764358+00:00"},{"alias_kind":"pith_short_12","alias_value":"PWIZRP5EM5AF","created_at":"2026-05-22T01:04:32.764358+00:00"},{"alias_kind":"pith_short_16","alias_value":"PWIZRP5EM5AFJ7CF","created_at":"2026-05-22T01:04:32.764358+00:00"},{"alias_kind":"pith_short_8","alias_value":"PWIZRP5E","created_at":"2026-05-22T01:04:32.764358+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/PWIZRP5EM5AFJ7CFQ4XY77VFDL","json":"https://pith.science/pith/PWIZRP5EM5AFJ7CFQ4XY77VFDL.json","graph_json":"https://pith.science/api/pith-number/PWIZRP5EM5AFJ7CFQ4XY77VFDL/graph.json","events_json":"https://pith.science/api/pith-number/PWIZRP5EM5AFJ7CFQ4XY77VFDL/events.json","paper":"https://pith.science/paper/PWIZRP5E"},"agent_actions":{"view_html":"https://pith.science/pith/PWIZRP5EM5AFJ7CFQ4XY77VFDL","download_json":"https://pith.science/pith/PWIZRP5EM5AFJ7CFQ4XY77VFDL.json","view_paper":"https://pith.science/paper/PWIZRP5E","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.22219&json=true","fetch_graph":"https://pith.science/api/pith-number/PWIZRP5EM5AFJ7CFQ4XY77VFDL/graph.json","fetch_events":"https://pith.science/api/pith-number/PWIZRP5EM5AFJ7CFQ4XY77VFDL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PWIZRP5EM5AFJ7CFQ4XY77VFDL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PWIZRP5EM5AFJ7CFQ4XY77VFDL/action/storage_attestation","attest_author":"https://pith.science/pith/PWIZRP5EM5AFJ7CFQ4XY77VFDL/action/author_attestation","sign_citation":"https://pith.science/pith/PWIZRP5EM5AFJ7CFQ4XY77VFDL/action/citation_signature","submit_replication":"https://pith.science/pith/PWIZRP5EM5AFJ7CFQ4XY77VFDL/action/replication_record"}},"created_at":"2026-05-22T01:04:32.764358+00:00","updated_at":"2026-05-22T01:04:32.764358+00:00"}