{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:5PXPYO667KU6FDRXOHTEN4I74B","short_pith_number":"pith:5PXPYO66","canonical_record":{"source":{"id":"2509.17455","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-22T07:49:58Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"8582c988452fe0f1f51afa344a26c76cc79c0065014035341ce08e05db1d30ee","abstract_canon_sha256":"666582238f3ff4e886f32eec5e76013ca3a690a53eee531e9c9093650fe64d84"},"schema_version":"1.0"},"canonical_sha256":"ebeefc3bdefaa9e28e3771e646f11fe077d4a79ca37fed7f06f6f4ea3bc8f801","source":{"kind":"arxiv","id":"2509.17455","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.17455","created_at":"2026-06-09T02:07:09Z"},{"alias_kind":"arxiv_version","alias_value":"2509.17455v2","created_at":"2026-06-09T02:07:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.17455","created_at":"2026-06-09T02:07:09Z"},{"alias_kind":"pith_short_12","alias_value":"5PXPYO667KU6","created_at":"2026-06-09T02:07:09Z"},{"alias_kind":"pith_short_16","alias_value":"5PXPYO667KU6FDRX","created_at":"2026-06-09T02:07:09Z"},{"alias_kind":"pith_short_8","alias_value":"5PXPYO66","created_at":"2026-06-09T02:07:09Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:5PXPYO667KU6FDRXOHTEN4I74B","target":"record","payload":{"canonical_record":{"source":{"id":"2509.17455","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-22T07:49:58Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"8582c988452fe0f1f51afa344a26c76cc79c0065014035341ce08e05db1d30ee","abstract_canon_sha256":"666582238f3ff4e886f32eec5e76013ca3a690a53eee531e9c9093650fe64d84"},"schema_version":"1.0"},"canonical_sha256":"ebeefc3bdefaa9e28e3771e646f11fe077d4a79ca37fed7f06f6f4ea3bc8f801","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T02:07:09.225757Z","signature_b64":"CWHvEFyTVv2QPn8cW5QhG68XE7F6ybApjWYov1lhLfREAttKfuySvY2ghs+tQUU9aMurs+aGAcxuLMkYOIhLDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ebeefc3bdefaa9e28e3771e646f11fe077d4a79ca37fed7f06f6f4ea3bc8f801","last_reissued_at":"2026-06-09T02:07:09.224897Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T02:07:09.224897Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2509.17455","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":"2026-06-09T02:07:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"J4sOT0gz+5vpsSpGeyi+udqqeNBU+EUTmDZggF90qUXzFeTsSdJXF5KXHfKKG1BfnRlTceJRQW4cFpfByOKQAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T05:56:36.110653Z"},"content_sha256":"40c66347d0a6587ab04688a750fe1a596f74f2483a37930bcf75a3e3cdcddcdb","schema_version":"1.0","event_id":"sha256:40c66347d0a6587ab04688a750fe1a596f74f2483a37930bcf75a3e3cdcddcdb"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:5PXPYO667KU6FDRXOHTEN4I74B","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Understanding Benchmark Language Under Weakened Formal Semantics","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Haoyang Chen, Kumiko Tanaka-Ishii","submitted_at":"2025-09-22T07:49:58Z","abstract_excerpt":"State-of-the-art NLP benchmarks require interpretation of natural language that specifies conditions, procedures, and exceptions, often relying on implicit assumptions and external knowledge. Constructing complete semantic representations with proof-theoretic guarantees is frequently impractical at scale, and purely text-based reasoning offers limited means of inspection. This paper asks how much understanding of benchmark language can be achieved when formal semantic guarantees are weakened. We investigate this question by extracting computables: executable representations whose runtime behav"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.17455","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/2509.17455/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":"2026-06-09T02:07:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WBw7Rutem5yukM/9ycFwjPeZlQNqz0+zh5/3PexXd+LRG7zlhrCyopAqLuIo3LzM4x0qzaKJ6uuq89e490cpCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T05:56:36.111248Z"},"content_sha256":"a3c7940432db1937751e6ec883b2a5e331eac05e55bd092c846bdd6414189f1f","schema_version":"1.0","event_id":"sha256:a3c7940432db1937751e6ec883b2a5e331eac05e55bd092c846bdd6414189f1f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5PXPYO667KU6FDRXOHTEN4I74B/bundle.json","state_url":"https://pith.science/pith/5PXPYO667KU6FDRXOHTEN4I74B/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5PXPYO667KU6FDRXOHTEN4I74B/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":"2026-08-18T05:56:36Z","links":{"resolver":"https://pith.science/pith/5PXPYO667KU6FDRXOHTEN4I74B","bundle":"https://pith.science/pith/5PXPYO667KU6FDRXOHTEN4I74B/bundle.json","state":"https://pith.science/pith/5PXPYO667KU6FDRXOHTEN4I74B/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5PXPYO667KU6FDRXOHTEN4I74B/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:5PXPYO667KU6FDRXOHTEN4I74B","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":"666582238f3ff4e886f32eec5e76013ca3a690a53eee531e9c9093650fe64d84","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-22T07:49:58Z","title_canon_sha256":"8582c988452fe0f1f51afa344a26c76cc79c0065014035341ce08e05db1d30ee"},"schema_version":"1.0","source":{"id":"2509.17455","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.17455","created_at":"2026-06-09T02:07:09Z"},{"alias_kind":"arxiv_version","alias_value":"2509.17455v2","created_at":"2026-06-09T02:07:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.17455","created_at":"2026-06-09T02:07:09Z"},{"alias_kind":"pith_short_12","alias_value":"5PXPYO667KU6","created_at":"2026-06-09T02:07:09Z"},{"alias_kind":"pith_short_16","alias_value":"5PXPYO667KU6FDRX","created_at":"2026-06-09T02:07:09Z"},{"alias_kind":"pith_short_8","alias_value":"5PXPYO66","created_at":"2026-06-09T02:07:09Z"}],"graph_snapshots":[{"event_id":"sha256:a3c7940432db1937751e6ec883b2a5e331eac05e55bd092c846bdd6414189f1f","target":"graph","created_at":"2026-06-09T02:07:09Z","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/2509.17455/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"State-of-the-art NLP benchmarks require interpretation of natural language that specifies conditions, procedures, and exceptions, often relying on implicit assumptions and external knowledge. Constructing complete semantic representations with proof-theoretic guarantees is frequently impractical at scale, and purely text-based reasoning offers limited means of inspection. This paper asks how much understanding of benchmark language can be achieved when formal semantic guarantees are weakened. We investigate this question by extracting computables: executable representations whose runtime behav","authors_text":"Haoyang Chen, Kumiko Tanaka-Ishii","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-22T07:49:58Z","title":"Understanding Benchmark Language Under Weakened Formal Semantics"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.17455","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":"sha256:40c66347d0a6587ab04688a750fe1a596f74f2483a37930bcf75a3e3cdcddcdb","target":"record","created_at":"2026-06-09T02:07:09Z","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":"666582238f3ff4e886f32eec5e76013ca3a690a53eee531e9c9093650fe64d84","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-22T07:49:58Z","title_canon_sha256":"8582c988452fe0f1f51afa344a26c76cc79c0065014035341ce08e05db1d30ee"},"schema_version":"1.0","source":{"id":"2509.17455","kind":"arxiv","version":2}},"canonical_sha256":"ebeefc3bdefaa9e28e3771e646f11fe077d4a79ca37fed7f06f6f4ea3bc8f801","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ebeefc3bdefaa9e28e3771e646f11fe077d4a79ca37fed7f06f6f4ea3bc8f801","first_computed_at":"2026-06-09T02:07:09.224897Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-09T02:07:09.224897Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"CWHvEFyTVv2QPn8cW5QhG68XE7F6ybApjWYov1lhLfREAttKfuySvY2ghs+tQUU9aMurs+aGAcxuLMkYOIhLDg==","signature_status":"signed_v1","signed_at":"2026-06-09T02:07:09.225757Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.17455","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:40c66347d0a6587ab04688a750fe1a596f74f2483a37930bcf75a3e3cdcddcdb","sha256:a3c7940432db1937751e6ec883b2a5e331eac05e55bd092c846bdd6414189f1f"],"state_sha256":"1e714b8952887f40d7709edb5913f2f9b3d288a859303101527f918c1745aa47"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Y6RDtwm2POVrOLSY/lHJ/t63u/VaxMUOxtZJrIxRxQrXDSAgGRXuub6msD5ZrrML0DiG4KkzfiEQWNCckmWsDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-18T05:56:36.115937Z","bundle_sha256":"cee9a5bace6dd86b42aa3429cfeaed1abaf40fb6de0842e90c15c905e60f82a0"}}