{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:NUGS44T4JFBN6KU3JVULLLGF4J","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"d53f093dec5e692652523d9e215916756e6f808fb1204d29951bbe37cf72a58b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-11T03:04:32Z","title_canon_sha256":"c433ad667a82ce42c4ef0854a6800f83689c95b54a127fdbe49b5654bf9e88bc"},"schema_version":"1.0","source":{"id":"2606.12837","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.12837","created_at":"2026-06-12T01:08:53Z"},{"alias_kind":"arxiv_version","alias_value":"2606.12837v1","created_at":"2026-06-12T01:08:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.12837","created_at":"2026-06-12T01:08:53Z"},{"alias_kind":"pith_short_12","alias_value":"NUGS44T4JFBN","created_at":"2026-06-12T01:08:53Z"},{"alias_kind":"pith_short_16","alias_value":"NUGS44T4JFBN6KU3","created_at":"2026-06-12T01:08:53Z"},{"alias_kind":"pith_short_8","alias_value":"NUGS44T4","created_at":"2026-06-12T01:08:53Z"}],"graph_snapshots":[{"event_id":"sha256:12d30c03efbe2fe0cd3d1dea3167993778a759cc1e8f80ef54af9bf29475ce77","target":"graph","created_at":"2026-06-12T01:08:53Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2606.12837/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Search agent benchmarks exemplified by BrowseComp have rapidly saturated over the past year, with the strongest models surpassing 90% accuracy. Since these benchmarks are predominantly human-authored, annotators lack a global perspective on entity statistics and cannot systematically maximize search space size and structural complexity. This creates a difficulty ceiling that is hard to break. To address this, we introduce LoHoSearch (Long-Horizon Search Agents), a challenging benchmark comprising 544 human-verified questions across 11 domains. LoHoSearch is constructed via an automated pipelin","authors_text":"Hao Yang, Jiarui Zhao, Lingchuan Liu, Rongzhi Zhang, Xi Su, Xunliang Cai","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-11T03:04:32Z","title":"LoHoSearch: Benchmarking Long-Horizon Search Agents Beyond the Human Difficulty Ceiling"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.12837","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:59ed8d4d00900e0fc0a2b00997728a1c092542e8ecf3bbc348ea8872803e6f39","target":"record","created_at":"2026-06-12T01:08:53Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"d53f093dec5e692652523d9e215916756e6f808fb1204d29951bbe37cf72a58b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-11T03:04:32Z","title_canon_sha256":"c433ad667a82ce42c4ef0854a6800f83689c95b54a127fdbe49b5654bf9e88bc"},"schema_version":"1.0","source":{"id":"2606.12837","kind":"arxiv","version":1}},"canonical_sha256":"6d0d2e727c4942df2a9b4d68b5acc5e24207bae8e72f474c12cf2a1b15972f46","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6d0d2e727c4942df2a9b4d68b5acc5e24207bae8e72f474c12cf2a1b15972f46","first_computed_at":"2026-06-12T01:08:53.490445Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-12T01:08:53.490445Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"i93RLQF9YgmajoyJnjr8bYgJYgGQOTwq/vhxPXiYvbO/wY8WrCSXIxWQ23ChK7kFZmTSR3TUl6cWINcQmtu3DQ==","signature_status":"signed_v1","signed_at":"2026-06-12T01:08:53.491242Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.12837","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:59ed8d4d00900e0fc0a2b00997728a1c092542e8ecf3bbc348ea8872803e6f39","sha256:12d30c03efbe2fe0cd3d1dea3167993778a759cc1e8f80ef54af9bf29475ce77"],"state_sha256":"56dfacedb465a215a31a0ca5b8483d8cba97bf15d697bf87ef1b5486080cca40"}