{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:DYO6ZH7IQYMLD6BYGH6F72JDSD","short_pith_number":"pith:DYO6ZH7I","schema_version":"1.0","canonical_sha256":"1e1dec9fe88618b1f83831fc5fe92390ea7325b521b1de7c2a7047f32a1bb69a","source":{"kind":"arxiv","id":"2607.05382","version":1},"attestation_state":"computed","paper":{"title":"Search Beyond What Can Be Taught: Evolving the Knowledge Boundary in Agentic Visual Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Che Liu, Cong Wei, Fangzhen Lin, Haozhe Wang, Jiaming Liu, Jimmy Lin, Jinpeng Yu, Ping Nie, Ruihua Huang, Weijia Feng, Wenhu Chen","submitted_at":"2026-07-06T17:56:12Z","abstract_excerpt":"Visual generators excel at rendering, but they confidently fabricate what they do not know. User requests are unbounded, evolving, and deeply long-tailed: new characters, trending entities, post-cutoff events, and more. This world-knowledge bottleneck is structural: generators are trained on fixed corpora, but the visual world is open-ended. We construct SearchGen-20K and SearchGen-Bench, with 20,839 prompts spanning twelve failure categories and twenty-two domains, paired with a pre-executed multimodal SearchGen-Corpus-1M to support offline, reproducible research. On SearchGen-Bench, frontier"},"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":"2607.05382","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-06T17:56:12Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"70eec35baa9e129c18ab9fd815890f60444992ab05950b9e07c9625a5b40bfc3","abstract_canon_sha256":"2c0aa265bbef09e31855e158861076d06037d9f33a5f0deb4a1906f40bbf3525"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T03:19:25.573625Z","signature_b64":"cTe5omL/tV5GsTmuyRBrVRRx+CoZ8ikiKr5PBIGuleqXn2+PsGEgcB3sU8cvS8ezICgVwQDANlBIWsPsPOBwAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1e1dec9fe88618b1f83831fc5fe92390ea7325b521b1de7c2a7047f32a1bb69a","last_reissued_at":"2026-07-07T03:19:25.573207Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T03:19:25.573207Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Search Beyond What Can Be Taught: Evolving the Knowledge Boundary in Agentic Visual Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Che Liu, Cong Wei, Fangzhen Lin, Haozhe Wang, Jiaming Liu, Jimmy Lin, Jinpeng Yu, Ping Nie, Ruihua Huang, Weijia Feng, Wenhu Chen","submitted_at":"2026-07-06T17:56:12Z","abstract_excerpt":"Visual generators excel at rendering, but they confidently fabricate what they do not know. User requests are unbounded, evolving, and deeply long-tailed: new characters, trending entities, post-cutoff events, and more. This world-knowledge bottleneck is structural: generators are trained on fixed corpora, but the visual world is open-ended. We construct SearchGen-20K and SearchGen-Bench, with 20,839 prompts spanning twelve failure categories and twenty-two domains, paired with a pre-executed multimodal SearchGen-Corpus-1M to support offline, reproducible research. On SearchGen-Bench, frontier"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.05382","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/2607.05382/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":"2607.05382","created_at":"2026-07-07T03:19:25.573263+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.05382v1","created_at":"2026-07-07T03:19:25.573263+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.05382","created_at":"2026-07-07T03:19:25.573263+00:00"},{"alias_kind":"pith_short_12","alias_value":"DYO6ZH7IQYML","created_at":"2026-07-07T03:19:25.573263+00:00"},{"alias_kind":"pith_short_16","alias_value":"DYO6ZH7IQYMLD6BY","created_at":"2026-07-07T03:19:25.573263+00:00"},{"alias_kind":"pith_short_8","alias_value":"DYO6ZH7I","created_at":"2026-07-07T03:19:25.573263+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2608.04436","citing_title":"ToolArtist: Tool-Using Unified Multimodal Models for Agentic Image Generation","ref_index":28,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DYO6ZH7IQYMLD6BYGH6F72JDSD","json":"https://pith.science/pith/DYO6ZH7IQYMLD6BYGH6F72JDSD.json","graph_json":"https://pith.science/api/pith-number/DYO6ZH7IQYMLD6BYGH6F72JDSD/graph.json","events_json":"https://pith.science/api/pith-number/DYO6ZH7IQYMLD6BYGH6F72JDSD/events.json","paper":"https://pith.science/paper/DYO6ZH7I"},"agent_actions":{"view_html":"https://pith.science/pith/DYO6ZH7IQYMLD6BYGH6F72JDSD","download_json":"https://pith.science/pith/DYO6ZH7IQYMLD6BYGH6F72JDSD.json","view_paper":"https://pith.science/paper/DYO6ZH7I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.05382&json=true","fetch_graph":"https://pith.science/api/pith-number/DYO6ZH7IQYMLD6BYGH6F72JDSD/graph.json","fetch_events":"https://pith.science/api/pith-number/DYO6ZH7IQYMLD6BYGH6F72JDSD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DYO6ZH7IQYMLD6BYGH6F72JDSD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DYO6ZH7IQYMLD6BYGH6F72JDSD/action/storage_attestation","attest_author":"https://pith.science/pith/DYO6ZH7IQYMLD6BYGH6F72JDSD/action/author_attestation","sign_citation":"https://pith.science/pith/DYO6ZH7IQYMLD6BYGH6F72JDSD/action/citation_signature","submit_replication":"https://pith.science/pith/DYO6ZH7IQYMLD6BYGH6F72JDSD/action/replication_record"}},"created_at":"2026-07-07T03:19:25.573263+00:00","updated_at":"2026-07-07T03:19:25.573263+00:00"}