{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:UETF4ATR2KA53GTMTDZWSTIEU3","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":"02477d102694fe73bd8e78efb04d378f252e84a8985c65c2644dd0c3ac71ef4b","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-30T05:46:01Z","title_canon_sha256":"800a48d25adb77bf0a20d3c14ce3f8456bb618a0e5d69ec9b914796bec50b559"},"schema_version":"1.0","source":{"id":"2606.26101","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.26101","created_at":"2026-06-26T00:15:26Z"},{"alias_kind":"arxiv_version","alias_value":"2606.26101v1","created_at":"2026-06-26T00:15:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.26101","created_at":"2026-06-26T00:15:26Z"},{"alias_kind":"pith_short_12","alias_value":"UETF4ATR2KA5","created_at":"2026-06-26T00:15:26Z"},{"alias_kind":"pith_short_16","alias_value":"UETF4ATR2KA53GTM","created_at":"2026-06-26T00:15:26Z"},{"alias_kind":"pith_short_8","alias_value":"UETF4ATR","created_at":"2026-06-26T00:15:26Z"}],"graph_snapshots":[{"event_id":"sha256:26013601f4bfd19ce5b1008733ae6cfa0d87b81a57af98b02c5ff47128d67bed","target":"graph","created_at":"2026-06-26T00:15:26Z","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/2606.26101/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reliable evaluation of large language models should separate supported answering from unsupported guessing without conflating either with data contamination, prompt idiosyncrasy, or generic refusal behavior. We present a contamination-aware, multi-zone benchmark for measuring the transition from answerable knowledge to abstention-expected unknowns under frozen build-time labels. The benchmark contains 1,200 items across five domains, explicit abstention expectations, contamination-risk metadata, and dual parsing with an official strict parser plus a normalized robustness parser. We evaluate FL","authors_text":"Bowen Zhang, Haoyi Wu, Jian Wang, Renwei Meng, Shengan Yang, Xican Wang, Xuanyan Qiu","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-30T05:46:01Z","title":"Know2Guess: A Contamination-Aware Multi-Zone Benchmark for Knowledge-Boundary Evaluation in Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.26101","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":"sha256:3992241477d37d3042fe5930d8cad4d8c39e8854db449aed999d9cb263956962","target":"record","created_at":"2026-06-26T00:15:26Z","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":"02477d102694fe73bd8e78efb04d378f252e84a8985c65c2644dd0c3ac71ef4b","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-30T05:46:01Z","title_canon_sha256":"800a48d25adb77bf0a20d3c14ce3f8456bb618a0e5d69ec9b914796bec50b559"},"schema_version":"1.0","source":{"id":"2606.26101","kind":"arxiv","version":1}},"canonical_sha256":"a1265e0271d281dd9a6c98f3694d04a6e5adbc4d848cac33cbeb3719e70c198f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a1265e0271d281dd9a6c98f3694d04a6e5adbc4d848cac33cbeb3719e70c198f","first_computed_at":"2026-06-26T00:15:26.301438Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-26T00:15:26.301438Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"o/vIZOrt0Qa52+hD7uInxOFZ5Omm1KoWaY+FKZ6JnBSzopv4ybFucy017MvxPF538NPIjtMze+jqDIFmwwjeCQ==","signature_status":"signed_v1","signed_at":"2026-06-26T00:15:26.301977Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.26101","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3992241477d37d3042fe5930d8cad4d8c39e8854db449aed999d9cb263956962","sha256:26013601f4bfd19ce5b1008733ae6cfa0d87b81a57af98b02c5ff47128d67bed"],"state_sha256":"8db536ec843c576a8f3fd0a5233c1bcf1e2a9ed8add7987b23d07336f2d7329e"}