{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:4QIJSFVJ6HXB3UWHN7I7BY4P5G","short_pith_number":"pith:4QIJSFVJ","canonical_record":{"source":{"id":"2607.02668","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-02T18:00:39Z","cross_cats_sorted":[],"title_canon_sha256":"2c43bf83f8b6e39d0de7cf6da3f5f70408289a41ea6f78bfa57013eec641313b","abstract_canon_sha256":"3e4e62d4cea995a3cd4445b6d60b8b5f58754fdd3fd1ac3e002eaf7cfb794e5e"},"schema_version":"1.0"},"canonical_sha256":"e4109916a9f1ee1dd2c76fd1f0e38fe99f4b370ca477cbb96b23b56e84e1907a","source":{"kind":"arxiv","id":"2607.02668","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.02668","created_at":"2026-07-07T00:16:07Z"},{"alias_kind":"arxiv_version","alias_value":"2607.02668v1","created_at":"2026-07-07T00:16:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.02668","created_at":"2026-07-07T00:16:07Z"},{"alias_kind":"pith_short_12","alias_value":"4QIJSFVJ6HXB","created_at":"2026-07-07T00:16:07Z"},{"alias_kind":"pith_short_16","alias_value":"4QIJSFVJ6HXB3UWH","created_at":"2026-07-07T00:16:07Z"},{"alias_kind":"pith_short_8","alias_value":"4QIJSFVJ","created_at":"2026-07-07T00:16:07Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:4QIJSFVJ6HXB3UWHN7I7BY4P5G","target":"record","payload":{"canonical_record":{"source":{"id":"2607.02668","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-02T18:00:39Z","cross_cats_sorted":[],"title_canon_sha256":"2c43bf83f8b6e39d0de7cf6da3f5f70408289a41ea6f78bfa57013eec641313b","abstract_canon_sha256":"3e4e62d4cea995a3cd4445b6d60b8b5f58754fdd3fd1ac3e002eaf7cfb794e5e"},"schema_version":"1.0"},"canonical_sha256":"e4109916a9f1ee1dd2c76fd1f0e38fe99f4b370ca477cbb96b23b56e84e1907a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T00:16:07.060938Z","signature_b64":"aQ67RW5+Hozr9Zaf9VrGeLdp/2huxHRaJT+vibIEdMRJki3UvItvnJtzsHUU0VMjYcgBQ7itoXULuNT7/7DVDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e4109916a9f1ee1dd2c76fd1f0e38fe99f4b370ca477cbb96b23b56e84e1907a","last_reissued_at":"2026-07-07T00:16:07.060217Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T00:16:07.060217Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.02668","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":"2026-07-07T00:16:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"C8z1lkuoDmNDDyyfdtblHEbWI1nqeJPt/gUD+zt+0tXSVIT+6OJBl6Z8qpOCVx/laEkuvzwqpeLOFdDY0w7LDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-24T15:20:34.400598Z"},"content_sha256":"d0da570eed10cca70655649fddce107e4ab8ab20d7b5476b01076f0be7b8f909","schema_version":"1.0","event_id":"sha256:d0da570eed10cca70655649fddce107e4ab8ab20d7b5476b01076f0be7b8f909"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:4QIJSFVJ6HXB3UWHN7I7BY4P5G","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Improving LLMs via Validator-to-Generator Alignment","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Greg Durrett, Jocelyn Zhang, Juan Diego Rodriguez, Katrin Erk","submitted_at":"2026-07-02T18:00:39Z","abstract_excerpt":"Large language models are inconsistent: varying prompts or including unrelated information can lead to unexpected changes in model outputs. The generator-validator (G-V) gap is one manifestation of this phenomenon, where LLMs generate responses that they then deem as invalid if re-queried to validate them. In this work, we introduce a new formulation of G-V consistency that involves a principled correction for utterance frequency. Specifically, generators often assign low likelihood to valid strings simply because those strings are a priori unlikely, which makes naive notions of G-V consistenc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.02668","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/2607.02668/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":"2026-07-07T00:16:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qQr3e0tz8BkzvZae/oUjY3lG0L9HgRF9mZTYaviLmwVD+aOJfQ41+xGaf2+TLksE81jkRlf0PttlP0UWdb0gAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-24T15:20:34.400980Z"},"content_sha256":"fdaa0a01346bc0acb947c4218e655385c17886d456d847aaad686c728c4450ac","schema_version":"1.0","event_id":"sha256:fdaa0a01346bc0acb947c4218e655385c17886d456d847aaad686c728c4450ac"},{"event_type":"integrity_finding","subject_pith_number":"pith:2026:4QIJSFVJ6HXB3UWHN7I7BY4P5G","target":"integrity","payload":{"note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.1162/TACL\\_A\\_00536) was visible in the surrounding text but could not be confirmed against doi.org as printed.","snippet":"Clara Meister, Tiago Pimentel, Gian Wiher, and Ryan Cotterell. 2023. https://doi.org/10.1162/TACL\\_A\\_00536 Locally typical sampling . Trans. Assoc. Comput. Linguistics, 11:102--121","arxiv_id":"2607.02668","detector":"doi_compliance","evidence":{"ref_index":80,"verdict_class":"incontrovertible","resolved_title":null,"printed_excerpt":"Clara Meister, Tiago Pimentel, Gian Wiher, and Ryan Cotterell. 2023. https://doi.org/10.1162/TACL\\_A\\_00536 Locally typical sampling . Trans. Assoc. Comput. Linguistics, 11:102--121","reconstructed_doi":"10.1162/TACL\\_A\\_00536"},"severity":"advisory","ref_index":80,"audited_at":"2026-07-12T08:10:07.826869Z","event_type":"pith.integrity.v1","detected_doi":"10.1162/TACL\\_A\\_00536","detector_url":"https://pith.science/pith-integrity-protocol#doi_compliance","external_url":null,"finding_type":"recoverable_identifier","evidence_hash":"f4fcf7f520b627bbaa1935c270a6b486f2085e71775501c1295a8961544d5eb1","paper_version":1,"verdict_class":"incontrovertible","resolved_title":null,"detector_version":"1.0.0","detected_arxiv_id":null,"integrity_event_id":12555,"payload_sha256":"8cca8a78fd74a58c9b8c065c5d6507942975c401b0e2f8cee13f8901302dd6dd","signature_b64":"sWDbU/Mc2nku9xf0F6UrnssWmZZhoTq2FQtPhp/P22Rr8aPJyjZeubOxwRLZgjmJJD/Rown3npSI5b8ZmOI1Dw==","signing_key_id":"pith-v1-2026-05"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-12T08:12:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OAtCjnOrw5rAksDziXkZBY52zeU0XZt2kkS8tAY2fCzOOuX79weVN2PZFCBKc3jDhcyfgaS9h0oqW3YhsQaBCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-24T15:20:34.402232Z"},"content_sha256":"4ca096fbe78380a49759c3c14343723a57a968d656e5a24e145281c6c255a3c5","schema_version":"1.0","event_id":"sha256:4ca096fbe78380a49759c3c14343723a57a968d656e5a24e145281c6c255a3c5"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4QIJSFVJ6HXB3UWHN7I7BY4P5G/bundle.json","state_url":"https://pith.science/pith/4QIJSFVJ6HXB3UWHN7I7BY4P5G/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4QIJSFVJ6HXB3UWHN7I7BY4P5G/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":"2026-07-24T15:20:34Z","links":{"resolver":"https://pith.science/pith/4QIJSFVJ6HXB3UWHN7I7BY4P5G","bundle":"https://pith.science/pith/4QIJSFVJ6HXB3UWHN7I7BY4P5G/bundle.json","state":"https://pith.science/pith/4QIJSFVJ6HXB3UWHN7I7BY4P5G/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4QIJSFVJ6HXB3UWHN7I7BY4P5G/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:4QIJSFVJ6HXB3UWHN7I7BY4P5G","merge_version":"pith-open-graph-merge-v1","event_count":3,"valid_event_count":3,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"3e4e62d4cea995a3cd4445b6d60b8b5f58754fdd3fd1ac3e002eaf7cfb794e5e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-02T18:00:39Z","title_canon_sha256":"2c43bf83f8b6e39d0de7cf6da3f5f70408289a41ea6f78bfa57013eec641313b"},"schema_version":"1.0","source":{"id":"2607.02668","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.02668","created_at":"2026-07-07T00:16:07Z"},{"alias_kind":"arxiv_version","alias_value":"2607.02668v1","created_at":"2026-07-07T00:16:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.02668","created_at":"2026-07-07T00:16:07Z"},{"alias_kind":"pith_short_12","alias_value":"4QIJSFVJ6HXB","created_at":"2026-07-07T00:16:07Z"},{"alias_kind":"pith_short_16","alias_value":"4QIJSFVJ6HXB3UWH","created_at":"2026-07-07T00:16:07Z"},{"alias_kind":"pith_short_8","alias_value":"4QIJSFVJ","created_at":"2026-07-07T00:16:07Z"}],"graph_snapshots":[{"event_id":"sha256:fdaa0a01346bc0acb947c4218e655385c17886d456d847aaad686c728c4450ac","target":"graph","created_at":"2026-07-07T00:16:07Z","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/2607.02668/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models are inconsistent: varying prompts or including unrelated information can lead to unexpected changes in model outputs. The generator-validator (G-V) gap is one manifestation of this phenomenon, where LLMs generate responses that they then deem as invalid if re-queried to validate them. In this work, we introduce a new formulation of G-V consistency that involves a principled correction for utterance frequency. Specifically, generators often assign low likelihood to valid strings simply because those strings are a priori unlikely, which makes naive notions of G-V consistenc","authors_text":"Greg Durrett, Jocelyn Zhang, Juan Diego Rodriguez, Katrin Erk","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-02T18:00:39Z","title":"Improving LLMs via Validator-to-Generator Alignment"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.02668","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:d0da570eed10cca70655649fddce107e4ab8ab20d7b5476b01076f0be7b8f909","target":"record","created_at":"2026-07-07T00:16:07Z","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":"3e4e62d4cea995a3cd4445b6d60b8b5f58754fdd3fd1ac3e002eaf7cfb794e5e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-02T18:00:39Z","title_canon_sha256":"2c43bf83f8b6e39d0de7cf6da3f5f70408289a41ea6f78bfa57013eec641313b"},"schema_version":"1.0","source":{"id":"2607.02668","kind":"arxiv","version":1}},"canonical_sha256":"e4109916a9f1ee1dd2c76fd1f0e38fe99f4b370ca477cbb96b23b56e84e1907a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e4109916a9f1ee1dd2c76fd1f0e38fe99f4b370ca477cbb96b23b56e84e1907a","first_computed_at":"2026-07-07T00:16:07.060217Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-07T00:16:07.060217Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"aQ67RW5+Hozr9Zaf9VrGeLdp/2huxHRaJT+vibIEdMRJki3UvItvnJtzsHUU0VMjYcgBQ7itoXULuNT7/7DVDA==","signature_status":"signed_v1","signed_at":"2026-07-07T00:16:07.060938Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.02668","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d0da570eed10cca70655649fddce107e4ab8ab20d7b5476b01076f0be7b8f909","sha256:fdaa0a01346bc0acb947c4218e655385c17886d456d847aaad686c728c4450ac","sha256:4ca096fbe78380a49759c3c14343723a57a968d656e5a24e145281c6c255a3c5"],"state_sha256":"29efc3251e26bae27a0b1668cf99efb43465f966b356204e04ec50794942a36e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nLHULtiaJu2Rld7651kabYn8n2dl9NsDGfTKbx9SZq0A2KhM/VYsHRoweGsivB9IOj9QYB02AkPGfDbkWWBmAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-24T15:20:34.403607Z","bundle_sha256":"4aa618dbdc3daef94c80160f9146a2469d023808d6758f2f197cdf246454b31b"}}