{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:GCZX6QCALGEZZ6WFG34B57O5GK","short_pith_number":"pith:GCZX6QCA","canonical_record":{"source":{"id":"2606.09450","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-08T12:57:18Z","cross_cats_sorted":[],"title_canon_sha256":"00a27cd2fceb5ecdd8fb407be1f5de039320f316fd734c2ba4605df33fcbd246","abstract_canon_sha256":"60012a3dbfcdd7220bc8d81ab43cbcc25e74f1cfe9de9b3cd3c67fddfeb86ed7"},"schema_version":"1.0"},"canonical_sha256":"30b37f404059899cfac536f81efddd328b9142b80f37b14327172dbd3642044b","source":{"kind":"arxiv","id":"2606.09450","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.09450","created_at":"2026-06-09T02:08:49Z"},{"alias_kind":"arxiv_version","alias_value":"2606.09450v1","created_at":"2026-06-09T02:08:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.09450","created_at":"2026-06-09T02:08:49Z"},{"alias_kind":"pith_short_12","alias_value":"GCZX6QCALGEZ","created_at":"2026-06-09T02:08:49Z"},{"alias_kind":"pith_short_16","alias_value":"GCZX6QCALGEZZ6WF","created_at":"2026-06-09T02:08:49Z"},{"alias_kind":"pith_short_8","alias_value":"GCZX6QCA","created_at":"2026-06-09T02:08:49Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:GCZX6QCALGEZZ6WFG34B57O5GK","target":"record","payload":{"canonical_record":{"source":{"id":"2606.09450","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-08T12:57:18Z","cross_cats_sorted":[],"title_canon_sha256":"00a27cd2fceb5ecdd8fb407be1f5de039320f316fd734c2ba4605df33fcbd246","abstract_canon_sha256":"60012a3dbfcdd7220bc8d81ab43cbcc25e74f1cfe9de9b3cd3c67fddfeb86ed7"},"schema_version":"1.0"},"canonical_sha256":"30b37f404059899cfac536f81efddd328b9142b80f37b14327172dbd3642044b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T02:08:49.254217Z","signature_b64":"RGZFldQ7QW6TP3H6UiFiE64nkJJ6slIWzzH93AVdHCCGtslWvKeD9f6PrEt6lCfgFsfCRh6I/V/hT8B5Oci8Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"30b37f404059899cfac536f81efddd328b9142b80f37b14327172dbd3642044b","last_reissued_at":"2026-06-09T02:08:49.253291Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T02:08:49.253291Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.09450","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":"2026-06-09T02:08:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CRQ8xSJTYDClY1iof12uXqwETfs30QgdHOrn+ekLiWR3xzqrfsn+vTnCXWUyj4WkR0ePWk7xa3Rhy4ryaNMbAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-03T16:47:41.993961Z"},"content_sha256":"374e2a9804915141019e4b6a42ba12bcb8360b197d544ef2d692640dc243d545","schema_version":"1.0","event_id":"sha256:374e2a9804915141019e4b6a42ba12bcb8360b197d544ef2d692640dc243d545"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:GCZX6QCALGEZZ6WFG34B57O5GK","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"TheoremBench: Evaluating LLMs on Theorem Proving in Formal Mathematics","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Andrey Galichin, Elvir Karimov, Ivan Oseledets, QuocViet Pham","submitted_at":"2026-06-08T12:57:18Z","abstract_excerpt":"LLMs have recently achieved strong results on formal proving benchmarks. However, existing evaluations remain heavily concentrated on competition-style problems and often fail to capture how models behave on longer, more dependency-rich mathematical developments. We introduce TheoremBench, a Lean4 benchmark designed to evaluate theorem provers beyond contest settings. The benchmark is built from nearly one hundred classical theorems and is released in two complementary forms: a plain main version containing one target theorem per instance, and a premised version that expands each theorem into "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.09450","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/2606.09450/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":"2026-06-09T02:08:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"M8IcqXOio2yx2rrAC6VwFNS+L9wNV6/t/vqsqs2bZyqbOFc03EKX6QpSgbAo9TlFJo4sEXzDhao60+GDzPQJAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-03T16:47:41.994354Z"},"content_sha256":"bbc7de4c0dbaba4da7b8f4330d688c7125a36686df2e00dc3e28565d85a58411","schema_version":"1.0","event_id":"sha256:bbc7de4c0dbaba4da7b8f4330d688c7125a36686df2e00dc3e28565d85a58411"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GCZX6QCALGEZZ6WFG34B57O5GK/bundle.json","state_url":"https://pith.science/pith/GCZX6QCALGEZZ6WFG34B57O5GK/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GCZX6QCALGEZZ6WFG34B57O5GK/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":"2026-07-03T16:47:41Z","links":{"resolver":"https://pith.science/pith/GCZX6QCALGEZZ6WFG34B57O5GK","bundle":"https://pith.science/pith/GCZX6QCALGEZZ6WFG34B57O5GK/bundle.json","state":"https://pith.science/pith/GCZX6QCALGEZZ6WFG34B57O5GK/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GCZX6QCALGEZZ6WFG34B57O5GK/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:GCZX6QCALGEZZ6WFG34B57O5GK","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":"60012a3dbfcdd7220bc8d81ab43cbcc25e74f1cfe9de9b3cd3c67fddfeb86ed7","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-08T12:57:18Z","title_canon_sha256":"00a27cd2fceb5ecdd8fb407be1f5de039320f316fd734c2ba4605df33fcbd246"},"schema_version":"1.0","source":{"id":"2606.09450","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.09450","created_at":"2026-06-09T02:08:49Z"},{"alias_kind":"arxiv_version","alias_value":"2606.09450v1","created_at":"2026-06-09T02:08:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.09450","created_at":"2026-06-09T02:08:49Z"},{"alias_kind":"pith_short_12","alias_value":"GCZX6QCALGEZ","created_at":"2026-06-09T02:08:49Z"},{"alias_kind":"pith_short_16","alias_value":"GCZX6QCALGEZZ6WF","created_at":"2026-06-09T02:08:49Z"},{"alias_kind":"pith_short_8","alias_value":"GCZX6QCA","created_at":"2026-06-09T02:08:49Z"}],"graph_snapshots":[{"event_id":"sha256:bbc7de4c0dbaba4da7b8f4330d688c7125a36686df2e00dc3e28565d85a58411","target":"graph","created_at":"2026-06-09T02:08:49Z","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/2606.09450/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"LLMs have recently achieved strong results on formal proving benchmarks. However, existing evaluations remain heavily concentrated on competition-style problems and often fail to capture how models behave on longer, more dependency-rich mathematical developments. We introduce TheoremBench, a Lean4 benchmark designed to evaluate theorem provers beyond contest settings. The benchmark is built from nearly one hundred classical theorems and is released in two complementary forms: a plain main version containing one target theorem per instance, and a premised version that expands each theorem into ","authors_text":"Andrey Galichin, Elvir Karimov, Ivan Oseledets, QuocViet Pham","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-08T12:57:18Z","title":"TheoremBench: Evaluating LLMs on Theorem Proving in Formal Mathematics"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.09450","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":"sha256:374e2a9804915141019e4b6a42ba12bcb8360b197d544ef2d692640dc243d545","target":"record","created_at":"2026-06-09T02:08:49Z","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":"60012a3dbfcdd7220bc8d81ab43cbcc25e74f1cfe9de9b3cd3c67fddfeb86ed7","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-08T12:57:18Z","title_canon_sha256":"00a27cd2fceb5ecdd8fb407be1f5de039320f316fd734c2ba4605df33fcbd246"},"schema_version":"1.0","source":{"id":"2606.09450","kind":"arxiv","version":1}},"canonical_sha256":"30b37f404059899cfac536f81efddd328b9142b80f37b14327172dbd3642044b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"30b37f404059899cfac536f81efddd328b9142b80f37b14327172dbd3642044b","first_computed_at":"2026-06-09T02:08:49.253291Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-09T02:08:49.253291Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"RGZFldQ7QW6TP3H6UiFiE64nkJJ6slIWzzH93AVdHCCGtslWvKeD9f6PrEt6lCfgFsfCRh6I/V/hT8B5Oci8Cw==","signature_status":"signed_v1","signed_at":"2026-06-09T02:08:49.254217Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.09450","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:374e2a9804915141019e4b6a42ba12bcb8360b197d544ef2d692640dc243d545","sha256:bbc7de4c0dbaba4da7b8f4330d688c7125a36686df2e00dc3e28565d85a58411"],"state_sha256":"86f2e46b798dfdb7e29653dbe7a09a7a9bc3538de6a22d68412633ea50ce65a7"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YaMKwUQREU7lpZ8WFl4Mt0lnuOJC996RGJ7UBrGNz1GhwbnNO0LorXoVKZrFcdYZCTVFSOnGJzpSSO9UHrH5Cg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-03T16:47:41.996525Z","bundle_sha256":"7a698e93f1289e74799b89dcce544c1b9ad310f28d6986fcb77132c578d2d271"}}