{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:PJCCLG6B373RNQ4562PEN3RJ5H","short_pith_number":"pith:PJCCLG6B","canonical_record":{"source":{"id":"2606.10799","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-09T12:46:35Z","cross_cats_sorted":[],"title_canon_sha256":"367ce6a0a5dcad459c07c59b9560f20ba08b4271a14616279f973f281c84b25c","abstract_canon_sha256":"064fc10284ceb5396ab9f9a6a1489f3fa67342db15d3762e070a3458e78b525d"},"schema_version":"1.0"},"canonical_sha256":"7a44259bc1dff716c39df69e46ee29e9f48fed688cfcc405792c16c7e0d6bd53","source":{"kind":"arxiv","id":"2606.10799","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.10799","created_at":"2026-06-10T01:10:41Z"},{"alias_kind":"arxiv_version","alias_value":"2606.10799v1","created_at":"2026-06-10T01:10:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.10799","created_at":"2026-06-10T01:10:41Z"},{"alias_kind":"pith_short_12","alias_value":"PJCCLG6B373R","created_at":"2026-06-10T01:10:41Z"},{"alias_kind":"pith_short_16","alias_value":"PJCCLG6B373RNQ45","created_at":"2026-06-10T01:10:41Z"},{"alias_kind":"pith_short_8","alias_value":"PJCCLG6B","created_at":"2026-06-10T01:10:41Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:PJCCLG6B373RNQ4562PEN3RJ5H","target":"record","payload":{"canonical_record":{"source":{"id":"2606.10799","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-09T12:46:35Z","cross_cats_sorted":[],"title_canon_sha256":"367ce6a0a5dcad459c07c59b9560f20ba08b4271a14616279f973f281c84b25c","abstract_canon_sha256":"064fc10284ceb5396ab9f9a6a1489f3fa67342db15d3762e070a3458e78b525d"},"schema_version":"1.0"},"canonical_sha256":"7a44259bc1dff716c39df69e46ee29e9f48fed688cfcc405792c16c7e0d6bd53","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-10T01:10:41.166084Z","signature_b64":"qStLAGz/+gTWQOL4Axf/7FfXpMWZBSPmVwshjBzqyjTMidwo9I4I+c1hav5zUQ4PYPfMOnzv86lz7/bcSGzlDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7a44259bc1dff716c39df69e46ee29e9f48fed688cfcc405792c16c7e0d6bd53","last_reissued_at":"2026-06-10T01:10:41.165626Z","signature_status":"signed_v1","first_computed_at":"2026-06-10T01:10:41.165626Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.10799","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":"2026-06-10T01:10:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yA66zAhtVO0t4LBjmJEvxBqnsfRS6q9ZEZT/Z3Apnr2Y2GdTj2ruS3BVNnqdWtOeEtG8jx+oG7aaWpGDR6DSAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-05T02:45:22.410577Z"},"content_sha256":"d4aaca4d1836b679572f11abbc57f3e7e2313d1402a9261f2680f7387f1f0b65","schema_version":"1.0","event_id":"sha256:d4aaca4d1836b679572f11abbc57f3e7e2313d1402a9261f2680f7387f1f0b65"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:PJCCLG6B373RNQ4562PEN3RJ5H","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Evaluating Research-Level Math Proofs via Strict Step-Level Verification","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Yifeng Sun","submitted_at":"2026-06-09T12:46:35Z","abstract_excerpt":"Large Language Models (LLMs) struggle to rigorously verify complex mathematical proofs. Standard global evaluation approaches suffer from \"context poisoning,\" in which superficially plausible statements mask subtle logical flaws, leading to hallucination or over-skepticism. To address this, we shift from global evaluation to strict step-level verification: our framework maintains detailed context for each deduction step and strictly constrains the sources of applied theorems. We evaluate on a carefully curated adversarial diagnostic suite of research-level proofs drawn from the FirstProof chal"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.10799","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/2606.10799/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":"2026-06-10T01:10:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CFBfSW9MioSMm/e3AQpoSirM1ryX+iTufkaDFrXbIBxvIpn3/GpM2BS2sOSZEbqto+9b2KpPtEizpYJJdWg/AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-05T02:45:22.410929Z"},"content_sha256":"858200551595b6a2a9b992eaddc3f43949e49957f8072d7ef0db4495a9b8c8bf","schema_version":"1.0","event_id":"sha256:858200551595b6a2a9b992eaddc3f43949e49957f8072d7ef0db4495a9b8c8bf"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PJCCLG6B373RNQ4562PEN3RJ5H/bundle.json","state_url":"https://pith.science/pith/PJCCLG6B373RNQ4562PEN3RJ5H/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PJCCLG6B373RNQ4562PEN3RJ5H/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":"2026-07-05T02:45:22Z","links":{"resolver":"https://pith.science/pith/PJCCLG6B373RNQ4562PEN3RJ5H","bundle":"https://pith.science/pith/PJCCLG6B373RNQ4562PEN3RJ5H/bundle.json","state":"https://pith.science/pith/PJCCLG6B373RNQ4562PEN3RJ5H/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PJCCLG6B373RNQ4562PEN3RJ5H/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:PJCCLG6B373RNQ4562PEN3RJ5H","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":"064fc10284ceb5396ab9f9a6a1489f3fa67342db15d3762e070a3458e78b525d","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-09T12:46:35Z","title_canon_sha256":"367ce6a0a5dcad459c07c59b9560f20ba08b4271a14616279f973f281c84b25c"},"schema_version":"1.0","source":{"id":"2606.10799","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.10799","created_at":"2026-06-10T01:10:41Z"},{"alias_kind":"arxiv_version","alias_value":"2606.10799v1","created_at":"2026-06-10T01:10:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.10799","created_at":"2026-06-10T01:10:41Z"},{"alias_kind":"pith_short_12","alias_value":"PJCCLG6B373R","created_at":"2026-06-10T01:10:41Z"},{"alias_kind":"pith_short_16","alias_value":"PJCCLG6B373RNQ45","created_at":"2026-06-10T01:10:41Z"},{"alias_kind":"pith_short_8","alias_value":"PJCCLG6B","created_at":"2026-06-10T01:10:41Z"}],"graph_snapshots":[{"event_id":"sha256:858200551595b6a2a9b992eaddc3f43949e49957f8072d7ef0db4495a9b8c8bf","target":"graph","created_at":"2026-06-10T01:10:41Z","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/2606.10799/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) struggle to rigorously verify complex mathematical proofs. Standard global evaluation approaches suffer from \"context poisoning,\" in which superficially plausible statements mask subtle logical flaws, leading to hallucination or over-skepticism. To address this, we shift from global evaluation to strict step-level verification: our framework maintains detailed context for each deduction step and strictly constrains the sources of applied theorems. We evaluate on a carefully curated adversarial diagnostic suite of research-level proofs drawn from the FirstProof chal","authors_text":"Yifeng Sun","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-09T12:46:35Z","title":"Evaluating Research-Level Math Proofs via Strict Step-Level Verification"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.10799","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:d4aaca4d1836b679572f11abbc57f3e7e2313d1402a9261f2680f7387f1f0b65","target":"record","created_at":"2026-06-10T01:10:41Z","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":"064fc10284ceb5396ab9f9a6a1489f3fa67342db15d3762e070a3458e78b525d","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-09T12:46:35Z","title_canon_sha256":"367ce6a0a5dcad459c07c59b9560f20ba08b4271a14616279f973f281c84b25c"},"schema_version":"1.0","source":{"id":"2606.10799","kind":"arxiv","version":1}},"canonical_sha256":"7a44259bc1dff716c39df69e46ee29e9f48fed688cfcc405792c16c7e0d6bd53","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7a44259bc1dff716c39df69e46ee29e9f48fed688cfcc405792c16c7e0d6bd53","first_computed_at":"2026-06-10T01:10:41.165626Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-10T01:10:41.165626Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"qStLAGz/+gTWQOL4Axf/7FfXpMWZBSPmVwshjBzqyjTMidwo9I4I+c1hav5zUQ4PYPfMOnzv86lz7/bcSGzlDQ==","signature_status":"signed_v1","signed_at":"2026-06-10T01:10:41.166084Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.10799","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d4aaca4d1836b679572f11abbc57f3e7e2313d1402a9261f2680f7387f1f0b65","sha256:858200551595b6a2a9b992eaddc3f43949e49957f8072d7ef0db4495a9b8c8bf"],"state_sha256":"f495b84a62fd01abad35d551733de497fdab149d9f2835da617048ecdfc2bc9a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"F9wOzjSG2tn17VIk7+ZdEeK3TuCMYGLVOEPkCsnFFSLS1neUxOdV4fit4kbGKoiw4qgc50VbY1nAjVf3cSj0CQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-05T02:45:22.412754Z","bundle_sha256":"647cca1be91e4da4d49c06f87bd784d49cb10d7a5bc41a420a342e0db5e9f96f"}}