{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:SZDEZMHJW7WQ4WCCDAL4FDIHZD","short_pith_number":"pith:SZDEZMHJ","canonical_record":{"source":{"id":"2602.16902","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-18T21:33:59Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"488149a5c5e7053d3acc6b5fded0803a333742ee5aff1001b9e03625616df387","abstract_canon_sha256":"48f51d3eb97548595d159b22e415adc76ad470930d992088446744f77ff61d73"},"schema_version":"1.0"},"canonical_sha256":"96464cb0e9b7ed0e58421817c28d07c8c2ca360c253e29f7ba44bc2835eb01a6","source":{"kind":"arxiv","id":"2602.16902","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"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:SZDEZMHJW7WQ4WCCDAL4FDIHZD","target":"record","payload":{"canonical_record":{"source":{"id":"2602.16902","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-18T21:33:59Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"488149a5c5e7053d3acc6b5fded0803a333742ee5aff1001b9e03625616df387","abstract_canon_sha256":"48f51d3eb97548595d159b22e415adc76ad470930d992088446744f77ff61d73"},"schema_version":"1.0"},"canonical_sha256":"96464cb0e9b7ed0e58421817c28d07c8c2ca360c253e29f7ba44bc2835eb01a6","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:03:43.800952Z","signature_b64":"N5R45SOjFy6KbVMrBoI8HvgmCuw1t6f0/iONJOjq+p2b3wHuP096j3FVAoHq/96gf+KcswCrGW6bfG3HyVfFBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"96464cb0e9b7ed0e58421817c28d07c8c2ca360c253e29f7ba44bc2835eb01a6","last_reissued_at":"2026-06-02T01:03:43.800461Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:03:43.800461Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2602.16902","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T01:03:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2Z4DWPiVRsIjh7Mzb0LN7i+iqpXSAsrYfz4g1injj7llk45LoY/40QfWJVaPmZiclbDWTCvOhQQQdavNpGxDBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-06T00:17:20.809810Z"},"content_sha256":"566b7590ca770e53fad9a7d31db5a99366410eebde48c1a2e6aa7b1718cb83c1","schema_version":"1.0","event_id":"sha256:566b7590ca770e53fad9a7d31db5a99366410eebde48c1a2e6aa7b1718cb83c1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:SZDEZMHJW7WQ4WCCDAL4FDIHZD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"LLM-WikiRace Benchmark: How Far Can LLMs Plan over Real-World Knowledge Graphs?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Ilija Bogunovic, Juliusz Ziomek, Lorenz Wolf, Shyam Sundhar Ramesh, William Bankes, Xiaohang Tang","submitted_at":"2026-02-18T21:33:59Z","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"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.16902","kind":"arxiv","version":4},"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/2602.16902/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T01:03:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PTMiJao135vouHZ/DrH6iN+F3PeKoU0OMfefXh/sfOSdo0TJOThg9UClRNSGHfUVuTFBOuhZOfeHky+LqeChAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-06T00:17:20.810471Z"},"content_sha256":"e93d9a56cfdfa987a303ae3f7ac2e8364106e1f38639597f84254bd9cb20e081","schema_version":"1.0","event_id":"sha256:e93d9a56cfdfa987a303ae3f7ac2e8364106e1f38639597f84254bd9cb20e081"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SZDEZMHJW7WQ4WCCDAL4FDIHZD/bundle.json","state_url":"https://pith.science/pith/SZDEZMHJW7WQ4WCCDAL4FDIHZD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SZDEZMHJW7WQ4WCCDAL4FDIHZD/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-06T00:17:20Z","links":{"resolver":"https://pith.science/pith/SZDEZMHJW7WQ4WCCDAL4FDIHZD","bundle":"https://pith.science/pith/SZDEZMHJW7WQ4WCCDAL4FDIHZD/bundle.json","state":"https://pith.science/pith/SZDEZMHJW7WQ4WCCDAL4FDIHZD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SZDEZMHJW7WQ4WCCDAL4FDIHZD/bundle.json"},"state":{"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"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7idxnV8g0oB3x3FKYlYj6NJqjAk/Zm+O68tD9cwYQq+5tc3KIMTCCmtl5QWeA2WRdgLGZAW+6Z/Tr5Wue68YBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-06T00:17:20.813723Z","bundle_sha256":"bc2f74d42c7d05085baf5e5a6db0674e02d078036558a822b8c2794b044b1a30"}}