{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:HUWW3WI4TKMJVYBEALNQ5GTXDG","short_pith_number":"pith:HUWW3WI4","canonical_record":{"source":{"id":"2507.16809","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-22T17:57:44Z","cross_cats_sorted":[],"title_canon_sha256":"2f1b039c33f4d4c2fbee9affc570f3205f0e2ce2bf7ba423aec5e15d8cd554e9","abstract_canon_sha256":"6258a1506eb9503307dfdffd41436b4842e18cd56a6859da9a726005bc5746c7"},"schema_version":"1.0"},"canonical_sha256":"3d2d6dd91c9a989ae02402db0e9a77199c5df37ccc23c5698647d8771a749e20","source":{"kind":"arxiv","id":"2507.16809","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.16809","created_at":"2026-07-05T11:42:33Z"},{"alias_kind":"arxiv_version","alias_value":"2507.16809v2","created_at":"2026-07-05T11:42:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.16809","created_at":"2026-07-05T11:42:33Z"},{"alias_kind":"pith_short_12","alias_value":"HUWW3WI4TKMJ","created_at":"2026-07-05T11:42:33Z"},{"alias_kind":"pith_short_16","alias_value":"HUWW3WI4TKMJVYBE","created_at":"2026-07-05T11:42:33Z"},{"alias_kind":"pith_short_8","alias_value":"HUWW3WI4","created_at":"2026-07-05T11:42:33Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:HUWW3WI4TKMJVYBEALNQ5GTXDG","target":"record","payload":{"canonical_record":{"source":{"id":"2507.16809","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-22T17:57:44Z","cross_cats_sorted":[],"title_canon_sha256":"2f1b039c33f4d4c2fbee9affc570f3205f0e2ce2bf7ba423aec5e15d8cd554e9","abstract_canon_sha256":"6258a1506eb9503307dfdffd41436b4842e18cd56a6859da9a726005bc5746c7"},"schema_version":"1.0"},"canonical_sha256":"3d2d6dd91c9a989ae02402db0e9a77199c5df37ccc23c5698647d8771a749e20","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:42:33.761957Z","signature_b64":"Xzj99bNTLio9yNWph58bFMLYoV7KWrhilSzgxTEEdtccbxBnHXyplGnICbrMF8+ILXbpBA6bY4fHnUOLLJhvCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3d2d6dd91c9a989ae02402db0e9a77199c5df37ccc23c5698647d8771a749e20","last_reissued_at":"2026-07-05T11:42:33.761371Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:42:33.761371Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.16809","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:42:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Fgo2vV+S/xNEXGQKnBBikpoilIZ5okp/enFkuj+RhNPxGGFr9UzZSF9ATyKV51Xwid0M0nLRWFIy2ZW64iRxCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T16:47:05.064455Z"},"content_sha256":"a5e57f55304b0607dfee905d6e644889215f0f0575ac59c58881764fcc69a666","schema_version":"1.0","event_id":"sha256:a5e57f55304b0607dfee905d6e644889215f0f0575ac59c58881764fcc69a666"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:HUWW3WI4TKMJVYBEALNQ5GTXDG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"LingBench++: A Linguistically-Informed Benchmark and Reasoning Framework for Multi-Step and Cross-Cultural Inference with LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chunki Lim, Da-Chen Lian, Guan-Yu Tseng, Pin-Cheng Chen, Pin-Er Chen, Ri-Sheng Huang, Shu-Kai Hsieh, You-Kuan Lin, Zhen-Yu Lin, Zi-Cheng Yang","submitted_at":"2025-07-22T17:57:44Z","abstract_excerpt":"We propose LingBench++, a linguistically-informed benchmark and reasoning framework designed to evaluate large language models (LLMs) on complex linguistic tasks inspired by the International Linguistics Olympiad (IOL). Unlike prior benchmarks that focus solely on final answer accuracy, LingBench++ provides structured reasoning traces, stepwise evaluation protocols, and rich typological metadata across over 90 low-resource and cross-cultural languages. We further develop a multi-agent architecture integrating grammatical knowledge retrieval, tool-augmented reasoning, and deliberate hypothesis "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.16809","kind":"arxiv","version":2},"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/2507.16809/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":"2026-07-05T11:42:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qo4kOselhR+q0MQNKtB88jP9wAT82mThonfiZsM1jSbgMdRkgk+GShzsqLTqbqlIlgTC85yn6DXOsiMgUFCiBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T16:47:05.065648Z"},"content_sha256":"7ec12fcd3caee02d5fcc70dcfceef6ec7911eb712ee9d806a17e024324efd947","schema_version":"1.0","event_id":"sha256:7ec12fcd3caee02d5fcc70dcfceef6ec7911eb712ee9d806a17e024324efd947"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HUWW3WI4TKMJVYBEALNQ5GTXDG/bundle.json","state_url":"https://pith.science/pith/HUWW3WI4TKMJVYBEALNQ5GTXDG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HUWW3WI4TKMJVYBEALNQ5GTXDG/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":"2026-08-03T16:47:05Z","links":{"resolver":"https://pith.science/pith/HUWW3WI4TKMJVYBEALNQ5GTXDG","bundle":"https://pith.science/pith/HUWW3WI4TKMJVYBEALNQ5GTXDG/bundle.json","state":"https://pith.science/pith/HUWW3WI4TKMJVYBEALNQ5GTXDG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HUWW3WI4TKMJVYBEALNQ5GTXDG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:HUWW3WI4TKMJVYBEALNQ5GTXDG","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":"6258a1506eb9503307dfdffd41436b4842e18cd56a6859da9a726005bc5746c7","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-22T17:57:44Z","title_canon_sha256":"2f1b039c33f4d4c2fbee9affc570f3205f0e2ce2bf7ba423aec5e15d8cd554e9"},"schema_version":"1.0","source":{"id":"2507.16809","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.16809","created_at":"2026-07-05T11:42:33Z"},{"alias_kind":"arxiv_version","alias_value":"2507.16809v2","created_at":"2026-07-05T11:42:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.16809","created_at":"2026-07-05T11:42:33Z"},{"alias_kind":"pith_short_12","alias_value":"HUWW3WI4TKMJ","created_at":"2026-07-05T11:42:33Z"},{"alias_kind":"pith_short_16","alias_value":"HUWW3WI4TKMJVYBE","created_at":"2026-07-05T11:42:33Z"},{"alias_kind":"pith_short_8","alias_value":"HUWW3WI4","created_at":"2026-07-05T11:42:33Z"}],"graph_snapshots":[{"event_id":"sha256:7ec12fcd3caee02d5fcc70dcfceef6ec7911eb712ee9d806a17e024324efd947","target":"graph","created_at":"2026-07-05T11:42:33Z","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/2507.16809/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We propose LingBench++, a linguistically-informed benchmark and reasoning framework designed to evaluate large language models (LLMs) on complex linguistic tasks inspired by the International Linguistics Olympiad (IOL). Unlike prior benchmarks that focus solely on final answer accuracy, LingBench++ provides structured reasoning traces, stepwise evaluation protocols, and rich typological metadata across over 90 low-resource and cross-cultural languages. We further develop a multi-agent architecture integrating grammatical knowledge retrieval, tool-augmented reasoning, and deliberate hypothesis ","authors_text":"Chunki Lim, Da-Chen Lian, Guan-Yu Tseng, Pin-Cheng Chen, Pin-Er Chen, Ri-Sheng Huang, Shu-Kai Hsieh, You-Kuan Lin, Zhen-Yu Lin, Zi-Cheng Yang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-22T17:57:44Z","title":"LingBench++: A Linguistically-Informed Benchmark and Reasoning Framework for Multi-Step and Cross-Cultural Inference with LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.16809","kind":"arxiv","version":2},"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:a5e57f55304b0607dfee905d6e644889215f0f0575ac59c58881764fcc69a666","target":"record","created_at":"2026-07-05T11:42:33Z","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":"6258a1506eb9503307dfdffd41436b4842e18cd56a6859da9a726005bc5746c7","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-22T17:57:44Z","title_canon_sha256":"2f1b039c33f4d4c2fbee9affc570f3205f0e2ce2bf7ba423aec5e15d8cd554e9"},"schema_version":"1.0","source":{"id":"2507.16809","kind":"arxiv","version":2}},"canonical_sha256":"3d2d6dd91c9a989ae02402db0e9a77199c5df37ccc23c5698647d8771a749e20","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3d2d6dd91c9a989ae02402db0e9a77199c5df37ccc23c5698647d8771a749e20","first_computed_at":"2026-07-05T11:42:33.761371Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:42:33.761371Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Xzj99bNTLio9yNWph58bFMLYoV7KWrhilSzgxTEEdtccbxBnHXyplGnICbrMF8+ILXbpBA6bY4fHnUOLLJhvCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:42:33.761957Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.16809","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a5e57f55304b0607dfee905d6e644889215f0f0575ac59c58881764fcc69a666","sha256:7ec12fcd3caee02d5fcc70dcfceef6ec7911eb712ee9d806a17e024324efd947"],"state_sha256":"bf747c20feb2f0a19ce55e4f40d3c0c3896ce05b49e8bb98f1cf0f12561f50d5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GUcw1WMVV5MDsrrnOoSpEiA6a3zi/3U/EjxgBlOwpWe6+SOw39Y5GM8RWzVdVj4A9va8KulOmJf6FJ5BTT2/BA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T16:47:05.083156Z","bundle_sha256":"a631b87839556dfcda9511e56d1ed990f7fc7e07f441a850d43b7952ec62e6af"}}