{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:QWKVSIJIJZV26TIQOZHXEVKTP4","short_pith_number":"pith:QWKVSIJI","canonical_record":{"source":{"id":"2602.20629","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-02-24T07:23:28Z","cross_cats_sorted":[],"title_canon_sha256":"e4e280e7dea72e0f3b2ae60119cedde4f4cf9d3cad35150ae7424b7ce3fbcb05","abstract_canon_sha256":"0af2978f753aadcf8d6db352ec67429c26d3365729347ad67076338175cdd464"},"schema_version":"1.0"},"canonical_sha256":"85955921284e6baf4d10764f7255537f12343582def81fb544aad4c14ba45041","source":{"kind":"arxiv","id":"2602.20629","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.20629","created_at":"2026-07-07T02:19:47Z"},{"alias_kind":"arxiv_version","alias_value":"2602.20629v3","created_at":"2026-07-07T02:19:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.20629","created_at":"2026-07-07T02:19:47Z"},{"alias_kind":"pith_short_12","alias_value":"QWKVSIJIJZV2","created_at":"2026-07-07T02:19:47Z"},{"alias_kind":"pith_short_16","alias_value":"QWKVSIJIJZV26TIQ","created_at":"2026-07-07T02:19:47Z"},{"alias_kind":"pith_short_8","alias_value":"QWKVSIJI","created_at":"2026-07-07T02:19:47Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:QWKVSIJIJZV26TIQOZHXEVKTP4","target":"record","payload":{"canonical_record":{"source":{"id":"2602.20629","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-02-24T07:23:28Z","cross_cats_sorted":[],"title_canon_sha256":"e4e280e7dea72e0f3b2ae60119cedde4f4cf9d3cad35150ae7424b7ce3fbcb05","abstract_canon_sha256":"0af2978f753aadcf8d6db352ec67429c26d3365729347ad67076338175cdd464"},"schema_version":"1.0"},"canonical_sha256":"85955921284e6baf4d10764f7255537f12343582def81fb544aad4c14ba45041","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T02:19:47.865454Z","signature_b64":"hds7Dwlu8x4fiEEDZdhbSi8c0PrBCFTRPRr/aWGrXkrAT5gHPI99/30qL6dt8Fc7uuAY+TlotP5IWQe7VkoEBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"85955921284e6baf4d10764f7255537f12343582def81fb544aad4c14ba45041","last_reissued_at":"2026-07-07T02:19:47.864485Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T02:19:47.864485Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2602.20629","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-07T02:19:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fUBYDVGiLXxKCesUfL5E87zVdSIqixgQ+u/EJDs6gLNsAyj3YdqGqGMCdYVjNCWQUp5mxtf5l7l/pOy1RC8pDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T21:15:18.710943Z"},"content_sha256":"3b3e86c10d5f95dfb8f6893524ba00751b616d77f611b84bfa902c6588bfbb4e","schema_version":"1.0","event_id":"sha256:3b3e86c10d5f95dfb8f6893524ba00751b616d77f611b84bfa902c6588bfbb4e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:QWKVSIJIJZV26TIQOZHXEVKTP4","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"QEDBENCH: Quantifying the Alignment Gap in Automated Evaluation of University-Level Mathematical Proofs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alan Goldfarb, Alireza Amiri Bavandpour, Ana Mickovic, Andres Miniguano-Trujillo, Annika Kanckos, Antoine Moulin, Ariane M. Masuda, Arman Cohan, Connor Stewart, Detchat Samart, Dhara Thakkar, Dino Lorenzini, Ebrahim Ghorbani, Edward Zhang, Giovanni Marzenta, Hossein Roodbarani, Hyungjun Choi, Hyun Jong Kim, Jeffrey Tse, Jing Guo, Ji Zeng, John Byrne, Jun Yan, Kevin Ren, Koppany Istv\\'an Encz, Levi Lorenzo, Lukas Mauth, Manjil Saikia, Mathieu Rundstr\\\"om, Meghal Gupta, Minjung Kang, Nujhat Ahmed Disha, Peter Ye, Polina Baron, Quanquan C. Liu, Rebecca Steiner, Robert Joseph George, Ruslans Aleksejevs, Santiago Gonzalez, Shubhrajit Bhattacharya, Sibel Yal\\c{c}{\\i}n, Stefano Huber, Sujeet Bhalerao, Tianyi Mao, Todor Anti\\'c, Tomos Parry, Vasiliki Velona, Wenqi Ni, Yuchen Fang, Yunhai Xiang, Zachary Burton","submitted_at":"2026-02-24T07:23:28Z","abstract_excerpt":"As Large Language Models (LLMs) saturate elementary benchmarks, the research frontier has shifted from generation to the reliability of automated evaluation. We demonstrate that standard \"LLM-as-a-Judge\" protocols suffer from a systematic Alignment Gap when applied to upper-undergraduate to early graduate level mathematics. To quantify this, we introduce QEDBench, the first large-scale dual-rubric alignment benchmark to systematically measure alignment with human experts on university-level math proofs by contrasting course-specific rubrics against expert common knowledge criteria. By deployin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.20629","kind":"arxiv","version":3},"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/2602.20629/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":"2026-07-07T02:19:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3IFIbExehOMlBaHYU8Ljeh2oRm96LMXfvMjHlA6FQGwj2EDE+mdxDiFgewekVy28IjIVz54ulfqjk3d+NjBFDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T21:15:18.711447Z"},"content_sha256":"0e9e6a48cabfb1c4d14ef4f24eb9cf66fd8f4797d6b1a9cd4692d1a0631e6f31","schema_version":"1.0","event_id":"sha256:0e9e6a48cabfb1c4d14ef4f24eb9cf66fd8f4797d6b1a9cd4692d1a0631e6f31"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QWKVSIJIJZV26TIQOZHXEVKTP4/bundle.json","state_url":"https://pith.science/pith/QWKVSIJIJZV26TIQOZHXEVKTP4/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QWKVSIJIJZV26TIQOZHXEVKTP4/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":"2026-08-19T21:15:18Z","links":{"resolver":"https://pith.science/pith/QWKVSIJIJZV26TIQOZHXEVKTP4","bundle":"https://pith.science/pith/QWKVSIJIJZV26TIQOZHXEVKTP4/bundle.json","state":"https://pith.science/pith/QWKVSIJIJZV26TIQOZHXEVKTP4/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QWKVSIJIJZV26TIQOZHXEVKTP4/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:QWKVSIJIJZV26TIQOZHXEVKTP4","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":"0af2978f753aadcf8d6db352ec67429c26d3365729347ad67076338175cdd464","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-02-24T07:23:28Z","title_canon_sha256":"e4e280e7dea72e0f3b2ae60119cedde4f4cf9d3cad35150ae7424b7ce3fbcb05"},"schema_version":"1.0","source":{"id":"2602.20629","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.20629","created_at":"2026-07-07T02:19:47Z"},{"alias_kind":"arxiv_version","alias_value":"2602.20629v3","created_at":"2026-07-07T02:19:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.20629","created_at":"2026-07-07T02:19:47Z"},{"alias_kind":"pith_short_12","alias_value":"QWKVSIJIJZV2","created_at":"2026-07-07T02:19:47Z"},{"alias_kind":"pith_short_16","alias_value":"QWKVSIJIJZV26TIQ","created_at":"2026-07-07T02:19:47Z"},{"alias_kind":"pith_short_8","alias_value":"QWKVSIJI","created_at":"2026-07-07T02:19:47Z"}],"graph_snapshots":[{"event_id":"sha256:0e9e6a48cabfb1c4d14ef4f24eb9cf66fd8f4797d6b1a9cd4692d1a0631e6f31","target":"graph","created_at":"2026-07-07T02:19:47Z","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/2602.20629/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As Large Language Models (LLMs) saturate elementary benchmarks, the research frontier has shifted from generation to the reliability of automated evaluation. We demonstrate that standard \"LLM-as-a-Judge\" protocols suffer from a systematic Alignment Gap when applied to upper-undergraduate to early graduate level mathematics. To quantify this, we introduce QEDBench, the first large-scale dual-rubric alignment benchmark to systematically measure alignment with human experts on university-level math proofs by contrasting course-specific rubrics against expert common knowledge criteria. By deployin","authors_text":"Alan Goldfarb, Alireza Amiri Bavandpour, Ana Mickovic, Andres Miniguano-Trujillo, Annika Kanckos, Antoine Moulin, Ariane M. Masuda, Arman Cohan, Connor Stewart, Detchat Samart, Dhara Thakkar, Dino Lorenzini, Ebrahim Ghorbani, Edward Zhang, Giovanni Marzenta, Hossein Roodbarani, Hyungjun Choi, Hyun Jong Kim, Jeffrey Tse, Jing Guo, Ji Zeng, John Byrne, Jun Yan, Kevin Ren, Koppany Istv\\'an Encz, Levi Lorenzo, Lukas Mauth, Manjil Saikia, Mathieu Rundstr\\\"om, Meghal Gupta, Minjung Kang, Nujhat Ahmed Disha, Peter Ye, Polina Baron, Quanquan C. Liu, Rebecca Steiner, Robert Joseph George, Ruslans Aleksejevs, Santiago Gonzalez, Shubhrajit Bhattacharya, Sibel Yal\\c{c}{\\i}n, Stefano Huber, Sujeet Bhalerao, Tianyi Mao, Todor Anti\\'c, Tomos Parry, Vasiliki Velona, Wenqi Ni, Yuchen Fang, Yunhai Xiang, Zachary Burton","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-02-24T07:23:28Z","title":"QEDBENCH: Quantifying the Alignment Gap in Automated Evaluation of University-Level Mathematical Proofs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.20629","kind":"arxiv","version":3},"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":"sha256:3b3e86c10d5f95dfb8f6893524ba00751b616d77f611b84bfa902c6588bfbb4e","target":"record","created_at":"2026-07-07T02:19:47Z","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":"0af2978f753aadcf8d6db352ec67429c26d3365729347ad67076338175cdd464","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-02-24T07:23:28Z","title_canon_sha256":"e4e280e7dea72e0f3b2ae60119cedde4f4cf9d3cad35150ae7424b7ce3fbcb05"},"schema_version":"1.0","source":{"id":"2602.20629","kind":"arxiv","version":3}},"canonical_sha256":"85955921284e6baf4d10764f7255537f12343582def81fb544aad4c14ba45041","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"85955921284e6baf4d10764f7255537f12343582def81fb544aad4c14ba45041","first_computed_at":"2026-07-07T02:19:47.864485Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-07T02:19:47.864485Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"hds7Dwlu8x4fiEEDZdhbSi8c0PrBCFTRPRr/aWGrXkrAT5gHPI99/30qL6dt8Fc7uuAY+TlotP5IWQe7VkoEBg==","signature_status":"signed_v1","signed_at":"2026-07-07T02:19:47.865454Z","signed_message":"canonical_sha256_bytes"},"source_id":"2602.20629","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3b3e86c10d5f95dfb8f6893524ba00751b616d77f611b84bfa902c6588bfbb4e","sha256:0e9e6a48cabfb1c4d14ef4f24eb9cf66fd8f4797d6b1a9cd4692d1a0631e6f31"],"state_sha256":"6ef58446d3c0b601aca0f81e3341d36e7f4dd36c223964f4c8756226ec1120ac"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uOzZEECa1Cw+r5Sq5vajdDF4njlLOa3AnB/gTQpzHLPzOv4dUZYqAsiF0NnhvRdAJlLVg2rGGPyDJvonSXvqBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-19T21:15:18.716239Z","bundle_sha256":"d6572c7230f8304b436276d92f99ae831e0b47cca78fb3c13902b7c338022978"}}