{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:SZDEZMHJW7WQ4WCCDAL4FDIHZD","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":"48f51d3eb97548595d159b22e415adc76ad470930d992088446744f77ff61d73","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-18T21:33:59Z","title_canon_sha256":"488149a5c5e7053d3acc6b5fded0803a333742ee5aff1001b9e03625616df387"},"schema_version":"1.0","source":{"id":"2602.16902","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.16902","created_at":"2026-06-02T01:03:43Z"},{"alias_kind":"arxiv_version","alias_value":"2602.16902v4","created_at":"2026-06-02T01:03:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.16902","created_at":"2026-06-02T01:03:43Z"},{"alias_kind":"pith_short_12","alias_value":"SZDEZMHJW7WQ","created_at":"2026-06-02T01:03:43Z"},{"alias_kind":"pith_short_16","alias_value":"SZDEZMHJW7WQ4WCC","created_at":"2026-06-02T01:03:43Z"},{"alias_kind":"pith_short_8","alias_value":"SZDEZMHJ","created_at":"2026-06-02T01:03:43Z"}],"graph_snapshots":[{"event_id":"sha256:e93d9a56cfdfa987a303ae3f7ac2e8364106e1f38639597f84254bd9cb20e081","target":"graph","created_at":"2026-06-02T01:03:43Z","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/2602.16902/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce LLM-Wikirace, a benchmark for evaluating planning, reasoning, and world knowledge in large language models (LLMs). In LLM-Wikirace, models must efficiently navigate Wikipedia hyperlinks step by step to reach a target page from a given source, requiring look-ahead planning and the ability to reason about how concepts are connected in the real world. We evaluate a broad set of open- and closed-source models, including Gemini-3, GPT-5, and Claude Opus 4.5, which achieve the strongest results on the easy level of the task and demonstrate superhuman performance. Despite this, performan","authors_text":"Ilija Bogunovic, Juliusz Ziomek, Lorenz Wolf, Shyam Sundhar Ramesh, William Bankes, Xiaohang Tang","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-18T21:33:59Z","title":"LLM-WikiRace Benchmark: How Far Can LLMs Plan over Real-World Knowledge Graphs?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.16902","kind":"arxiv","version":4},"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":"sha256:566b7590ca770e53fad9a7d31db5a99366410eebde48c1a2e6aa7b1718cb83c1","target":"record","created_at":"2026-06-02T01:03:43Z","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":"48f51d3eb97548595d159b22e415adc76ad470930d992088446744f77ff61d73","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-18T21:33:59Z","title_canon_sha256":"488149a5c5e7053d3acc6b5fded0803a333742ee5aff1001b9e03625616df387"},"schema_version":"1.0","source":{"id":"2602.16902","kind":"arxiv","version":4}},"canonical_sha256":"96464cb0e9b7ed0e58421817c28d07c8c2ca360c253e29f7ba44bc2835eb01a6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"96464cb0e9b7ed0e58421817c28d07c8c2ca360c253e29f7ba44bc2835eb01a6","first_computed_at":"2026-06-02T01:03:43.800461Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T01:03:43.800461Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"N5R45SOjFy6KbVMrBoI8HvgmCuw1t6f0/iONJOjq+p2b3wHuP096j3FVAoHq/96gf+KcswCrGW6bfG3HyVfFBw==","signature_status":"signed_v1","signed_at":"2026-06-02T01:03:43.800952Z","signed_message":"canonical_sha256_bytes"},"source_id":"2602.16902","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:566b7590ca770e53fad9a7d31db5a99366410eebde48c1a2e6aa7b1718cb83c1","sha256:e93d9a56cfdfa987a303ae3f7ac2e8364106e1f38639597f84254bd9cb20e081"],"state_sha256":"c3c31b5ab3b59829abe74acc58ad2f7168979b5dee4855040399b851091aff2f"}