{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:X4XLBOW5R4CMJH4J6GV554GN3B","short_pith_number":"pith:X4XLBOW5","canonical_record":{"source":{"id":"2405.06677","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-05T02:06:37Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2003b09d6a801594d3dacfd9bebfa1594a2893f1a759e14219c70f61e0742aec","abstract_canon_sha256":"7349ca7aed008283ccab03c9d9f18ed166a77f58e3ca41c5d3d05a2145a6000c"},"schema_version":"1.0"},"canonical_sha256":"bf2eb0badd8f04c49f89f1abdef0cdd85d106eeb5be96ad6b5f551a8aa946213","source":{"kind":"arxiv","id":"2405.06677","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.06677","created_at":"2026-07-05T08:18:10Z"},{"alias_kind":"arxiv_version","alias_value":"2405.06677v1","created_at":"2026-07-05T08:18:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.06677","created_at":"2026-07-05T08:18:10Z"},{"alias_kind":"pith_short_12","alias_value":"X4XLBOW5R4CM","created_at":"2026-07-05T08:18:10Z"},{"alias_kind":"pith_short_16","alias_value":"X4XLBOW5R4CMJH4J","created_at":"2026-07-05T08:18:10Z"},{"alias_kind":"pith_short_8","alias_value":"X4XLBOW5","created_at":"2026-07-05T08:18:10Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:X4XLBOW5R4CMJH4J6GV554GN3B","target":"record","payload":{"canonical_record":{"source":{"id":"2405.06677","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-05T02:06:37Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2003b09d6a801594d3dacfd9bebfa1594a2893f1a759e14219c70f61e0742aec","abstract_canon_sha256":"7349ca7aed008283ccab03c9d9f18ed166a77f58e3ca41c5d3d05a2145a6000c"},"schema_version":"1.0"},"canonical_sha256":"bf2eb0badd8f04c49f89f1abdef0cdd85d106eeb5be96ad6b5f551a8aa946213","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:18:10.197864Z","signature_b64":"0gIoVD8sZqg/UvVteixGLYIoJ8Oi64FcppSgxOFUFUVw4KyFjjLEo/mB8bzhqonYVPNxQjcsT/EucWns/zpZBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bf2eb0badd8f04c49f89f1abdef0cdd85d106eeb5be96ad6b5f551a8aa946213","last_reissued_at":"2026-07-05T08:18:10.197342Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:18:10.197342Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2405.06677","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":"2026-07-05T08:18:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kXzfHUTkLs7JP0BymEtT5lhOkiuysSzxtq5jsElU7cXbmhjKWqbhaS6zSgrAbpezEkbNOR5HdFUAtaH0XGUgBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T22:50:24.347540Z"},"content_sha256":"a846c9b1a1b50f51ff590e131efdc334b099b9dc5ff83cbf644ea7858cedb4a7","schema_version":"1.0","event_id":"sha256:a846c9b1a1b50f51ff590e131efdc334b099b9dc5ff83cbf644ea7858cedb4a7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:X4XLBOW5R4CMJH4J6GV554GN3B","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ATG: Benchmarking Automated Theorem Generation for Generative Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Qingxing Cao, Xiaodan Liang, Xiaohan Lin, Yinya Huang, Zhenguo Li, Zhengying Liu, Zhicheng Yang","submitted_at":"2024-05-05T02:06:37Z","abstract_excerpt":"Humans can develop new theorems to explore broader and more complex mathematical results. While current generative language models (LMs) have achieved significant improvement in automatically proving theorems, their ability to generate new or reusable theorems is still under-explored. Without the new theorems, current LMs struggle to prove harder theorems that are distant from the given hypotheses with the exponentially growing search space. Therefore, this paper proposes an Automated Theorem Generation (ATG) benchmark that evaluates whether an agent can automatically generate valuable (and po"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.06677","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/2405.06677/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":"2026-07-05T08:18:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qUaWnTjfEmItb+rmjr2NY13PP3qzFmCYd8rCugB36jH0PleT55UxmJxCCJ0ZO/FdPV12rMHUzIcjoAKeIBxTDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T22:50:24.348460Z"},"content_sha256":"c6eb4431d86cbf6bf5aa31c5689b1c4e690c373524ddd351f71bb834be71511e","schema_version":"1.0","event_id":"sha256:c6eb4431d86cbf6bf5aa31c5689b1c4e690c373524ddd351f71bb834be71511e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/X4XLBOW5R4CMJH4J6GV554GN3B/bundle.json","state_url":"https://pith.science/pith/X4XLBOW5R4CMJH4J6GV554GN3B/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/X4XLBOW5R4CMJH4J6GV554GN3B/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":"2026-08-06T22:50:24Z","links":{"resolver":"https://pith.science/pith/X4XLBOW5R4CMJH4J6GV554GN3B","bundle":"https://pith.science/pith/X4XLBOW5R4CMJH4J6GV554GN3B/bundle.json","state":"https://pith.science/pith/X4XLBOW5R4CMJH4J6GV554GN3B/state.json","well_known_bundle":"https://pith.science/.well-known/pith/X4XLBOW5R4CMJH4J6GV554GN3B/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:X4XLBOW5R4CMJH4J6GV554GN3B","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":"7349ca7aed008283ccab03c9d9f18ed166a77f58e3ca41c5d3d05a2145a6000c","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-05T02:06:37Z","title_canon_sha256":"2003b09d6a801594d3dacfd9bebfa1594a2893f1a759e14219c70f61e0742aec"},"schema_version":"1.0","source":{"id":"2405.06677","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.06677","created_at":"2026-07-05T08:18:10Z"},{"alias_kind":"arxiv_version","alias_value":"2405.06677v1","created_at":"2026-07-05T08:18:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.06677","created_at":"2026-07-05T08:18:10Z"},{"alias_kind":"pith_short_12","alias_value":"X4XLBOW5R4CM","created_at":"2026-07-05T08:18:10Z"},{"alias_kind":"pith_short_16","alias_value":"X4XLBOW5R4CMJH4J","created_at":"2026-07-05T08:18:10Z"},{"alias_kind":"pith_short_8","alias_value":"X4XLBOW5","created_at":"2026-07-05T08:18:10Z"}],"graph_snapshots":[{"event_id":"sha256:c6eb4431d86cbf6bf5aa31c5689b1c4e690c373524ddd351f71bb834be71511e","target":"graph","created_at":"2026-07-05T08:18:10Z","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/2405.06677/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Humans can develop new theorems to explore broader and more complex mathematical results. While current generative language models (LMs) have achieved significant improvement in automatically proving theorems, their ability to generate new or reusable theorems is still under-explored. Without the new theorems, current LMs struggle to prove harder theorems that are distant from the given hypotheses with the exponentially growing search space. Therefore, this paper proposes an Automated Theorem Generation (ATG) benchmark that evaluates whether an agent can automatically generate valuable (and po","authors_text":"Qingxing Cao, Xiaodan Liang, Xiaohan Lin, Yinya Huang, Zhenguo Li, Zhengying Liu, Zhicheng Yang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-05T02:06:37Z","title":"ATG: Benchmarking Automated Theorem Generation for Generative Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.06677","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:a846c9b1a1b50f51ff590e131efdc334b099b9dc5ff83cbf644ea7858cedb4a7","target":"record","created_at":"2026-07-05T08:18:10Z","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":"7349ca7aed008283ccab03c9d9f18ed166a77f58e3ca41c5d3d05a2145a6000c","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-05T02:06:37Z","title_canon_sha256":"2003b09d6a801594d3dacfd9bebfa1594a2893f1a759e14219c70f61e0742aec"},"schema_version":"1.0","source":{"id":"2405.06677","kind":"arxiv","version":1}},"canonical_sha256":"bf2eb0badd8f04c49f89f1abdef0cdd85d106eeb5be96ad6b5f551a8aa946213","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bf2eb0badd8f04c49f89f1abdef0cdd85d106eeb5be96ad6b5f551a8aa946213","first_computed_at":"2026-07-05T08:18:10.197342Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:18:10.197342Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"0gIoVD8sZqg/UvVteixGLYIoJ8Oi64FcppSgxOFUFUVw4KyFjjLEo/mB8bzhqonYVPNxQjcsT/EucWns/zpZBw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:18:10.197864Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.06677","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a846c9b1a1b50f51ff590e131efdc334b099b9dc5ff83cbf644ea7858cedb4a7","sha256:c6eb4431d86cbf6bf5aa31c5689b1c4e690c373524ddd351f71bb834be71511e"],"state_sha256":"57026ff62ab739d58c49442475b4f3247f682015bfd52ef4d3d2ea66f3ad2fef"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+UZBd/n/wqeux2Adw52rTD2ru8hnNEl8AVh0Q2n2ofeq7TdACeBij7F7vDmGQMVhSdPKAE5B1aW5rhaAITGwDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T22:50:24.353292Z","bundle_sha256":"f4bc5b236139da87142fc9e30c13a6e1dab842963e42428967280aa437b54ff5"}}