{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:Y6QLESE7CD7MIMIU53AYQP6N74","short_pith_number":"pith:Y6QLESE7","canonical_record":{"source":{"id":"2505.11855","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-17T05:45:16Z","cross_cats_sorted":[],"title_canon_sha256":"b0e01c82198b3d1fae630f781a2399ac6a809f8e8141d4d5048f9513e19008e8","abstract_canon_sha256":"d872a0a9e477ecfa02ee4f0d3757be25458e7b15388d38aba519b60c6e0305ab"},"schema_version":"1.0"},"canonical_sha256":"c7a0b2489f10fec43114eec1883fcdff148b3a25faf41f5a66fe67b37218ace7","source":{"kind":"arxiv","id":"2505.11855","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.11855","created_at":"2026-07-05T11:04:53Z"},{"alias_kind":"arxiv_version","alias_value":"2505.11855v1","created_at":"2026-07-05T11:04:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.11855","created_at":"2026-07-05T11:04:53Z"},{"alias_kind":"pith_short_12","alias_value":"Y6QLESE7CD7M","created_at":"2026-07-05T11:04:53Z"},{"alias_kind":"pith_short_16","alias_value":"Y6QLESE7CD7MIMIU","created_at":"2026-07-05T11:04:53Z"},{"alias_kind":"pith_short_8","alias_value":"Y6QLESE7","created_at":"2026-07-05T11:04:53Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:Y6QLESE7CD7MIMIU53AYQP6N74","target":"record","payload":{"canonical_record":{"source":{"id":"2505.11855","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-17T05:45:16Z","cross_cats_sorted":[],"title_canon_sha256":"b0e01c82198b3d1fae630f781a2399ac6a809f8e8141d4d5048f9513e19008e8","abstract_canon_sha256":"d872a0a9e477ecfa02ee4f0d3757be25458e7b15388d38aba519b60c6e0305ab"},"schema_version":"1.0"},"canonical_sha256":"c7a0b2489f10fec43114eec1883fcdff148b3a25faf41f5a66fe67b37218ace7","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:04:53.272994Z","signature_b64":"tcVsOShPKZrQMyIOa9O0qxKAcis8wR12YbL478/52Mpgp2KizpcOf0MvTfv2A6TTkomdIiQ4vSRUKM3Xt10JBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c7a0b2489f10fec43114eec1883fcdff148b3a25faf41f5a66fe67b37218ace7","last_reissued_at":"2026-07-05T11:04:53.272499Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:04:53.272499Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.11855","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:04:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xBaISUZ5f5+qYG34HE+uO+oR75+cRqUbQwRwof3AvtTvfslJv6RSTBaNh7zDqboUZ9jOaPDMWxpJUNltwwU6Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-08T13:46:32.675019Z"},"content_sha256":"885211f9ee7072da547e578614a779c5e09ca8a3289f10679548d99f6071cec1","schema_version":"1.0","event_id":"sha256:885211f9ee7072da547e578614a779c5e09ca8a3289f10679548d99f6071cec1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:Y6QLESE7CD7MIMIU53AYQP6N74","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"When AI Co-Scientists Fail: SPOT-a Benchmark for Automated Verification of Scientific Research","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Gon\\c{c}alo Paulo, Guijin Son, Heejeong Nam, Honglu Fan, Hyunwoo Ko, Jinha Choi, Jinyeop Song, Jiwoo Hong, Seungwon Lim, Stella Biderman, Youngjae Yu","submitted_at":"2025-05-17T05:45:16Z","abstract_excerpt":"Recent advances in large language models (LLMs) have fueled the vision of automated scientific discovery, often called AI Co-Scientists. To date, prior work casts these systems as generative co-authors responsible for crafting hypotheses, synthesizing code, or drafting manuscripts. In this work, we explore a complementary application: using LLMs as verifiers to automate the \\textbf{academic verification of scientific manuscripts}. To that end, we introduce SPOT, a dataset of 83 published papers paired with 91 errors significant enough to prompt errata or retraction, cross-validated with actual"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.11855","kind":"arxiv","version":1},"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/2505.11855/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":"2026-07-05T11:04:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DBBOBj8bKYFcwBnsHBETO9s/DIW+E0OtZZlABonV1/kEygLtUUZJ1j6DMlmcM8yukr9YUQhpJWCWeGmxPreZAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-08T13:46:32.675404Z"},"content_sha256":"9cd2d9a5f48a4825c2f570064948438c0bb2afe95f7f8adb9f6b3eb24160b6fd","schema_version":"1.0","event_id":"sha256:9cd2d9a5f48a4825c2f570064948438c0bb2afe95f7f8adb9f6b3eb24160b6fd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Y6QLESE7CD7MIMIU53AYQP6N74/bundle.json","state_url":"https://pith.science/pith/Y6QLESE7CD7MIMIU53AYQP6N74/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Y6QLESE7CD7MIMIU53AYQP6N74/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":"2026-07-08T13:46:32Z","links":{"resolver":"https://pith.science/pith/Y6QLESE7CD7MIMIU53AYQP6N74","bundle":"https://pith.science/pith/Y6QLESE7CD7MIMIU53AYQP6N74/bundle.json","state":"https://pith.science/pith/Y6QLESE7CD7MIMIU53AYQP6N74/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Y6QLESE7CD7MIMIU53AYQP6N74/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:Y6QLESE7CD7MIMIU53AYQP6N74","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":"d872a0a9e477ecfa02ee4f0d3757be25458e7b15388d38aba519b60c6e0305ab","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-17T05:45:16Z","title_canon_sha256":"b0e01c82198b3d1fae630f781a2399ac6a809f8e8141d4d5048f9513e19008e8"},"schema_version":"1.0","source":{"id":"2505.11855","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.11855","created_at":"2026-07-05T11:04:53Z"},{"alias_kind":"arxiv_version","alias_value":"2505.11855v1","created_at":"2026-07-05T11:04:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.11855","created_at":"2026-07-05T11:04:53Z"},{"alias_kind":"pith_short_12","alias_value":"Y6QLESE7CD7M","created_at":"2026-07-05T11:04:53Z"},{"alias_kind":"pith_short_16","alias_value":"Y6QLESE7CD7MIMIU","created_at":"2026-07-05T11:04:53Z"},{"alias_kind":"pith_short_8","alias_value":"Y6QLESE7","created_at":"2026-07-05T11:04:53Z"}],"graph_snapshots":[{"event_id":"sha256:9cd2d9a5f48a4825c2f570064948438c0bb2afe95f7f8adb9f6b3eb24160b6fd","target":"graph","created_at":"2026-07-05T11:04:53Z","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/2505.11855/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent advances in large language models (LLMs) have fueled the vision of automated scientific discovery, often called AI Co-Scientists. To date, prior work casts these systems as generative co-authors responsible for crafting hypotheses, synthesizing code, or drafting manuscripts. In this work, we explore a complementary application: using LLMs as verifiers to automate the \\textbf{academic verification of scientific manuscripts}. To that end, we introduce SPOT, a dataset of 83 published papers paired with 91 errors significant enough to prompt errata or retraction, cross-validated with actual","authors_text":"Gon\\c{c}alo Paulo, Guijin Son, Heejeong Nam, Honglu Fan, Hyunwoo Ko, Jinha Choi, Jinyeop Song, Jiwoo Hong, Seungwon Lim, Stella Biderman, Youngjae Yu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-17T05:45:16Z","title":"When AI Co-Scientists Fail: SPOT-a Benchmark for Automated Verification of Scientific Research"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.11855","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":"sha256:885211f9ee7072da547e578614a779c5e09ca8a3289f10679548d99f6071cec1","target":"record","created_at":"2026-07-05T11:04:53Z","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":"d872a0a9e477ecfa02ee4f0d3757be25458e7b15388d38aba519b60c6e0305ab","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-17T05:45:16Z","title_canon_sha256":"b0e01c82198b3d1fae630f781a2399ac6a809f8e8141d4d5048f9513e19008e8"},"schema_version":"1.0","source":{"id":"2505.11855","kind":"arxiv","version":1}},"canonical_sha256":"c7a0b2489f10fec43114eec1883fcdff148b3a25faf41f5a66fe67b37218ace7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c7a0b2489f10fec43114eec1883fcdff148b3a25faf41f5a66fe67b37218ace7","first_computed_at":"2026-07-05T11:04:53.272499Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:04:53.272499Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"tcVsOShPKZrQMyIOa9O0qxKAcis8wR12YbL478/52Mpgp2KizpcOf0MvTfv2A6TTkomdIiQ4vSRUKM3Xt10JBg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:04:53.272994Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.11855","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:885211f9ee7072da547e578614a779c5e09ca8a3289f10679548d99f6071cec1","sha256:9cd2d9a5f48a4825c2f570064948438c0bb2afe95f7f8adb9f6b3eb24160b6fd"],"state_sha256":"7b07266588f849c7337abe4321fe4a69a3fc69b9218afa7ab6d958eb46587cf4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qgKY4nQ5fwyvIszyo1NeM/LXcz60LzRoW/b1hcIZZXVEA7+vTg378vQX8fSxVpSsCrxCEBhDSgoyKWakM+7pAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-08T13:46:32.677559Z","bundle_sha256":"3aefb340739a0506411ad356bfff277f304fcb174cce673a39404618c6e05a1c"}}