{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:ZD56OO7JFXC3PRZ452D2HRASRD","short_pith_number":"pith:ZD56OO7J","canonical_record":{"source":{"id":"2607.17524","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-20T03:57:22Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"0ba5bf7123c7a5f13182c4c58ce60065149f6ee70f7ce8705173106426b4c498","abstract_canon_sha256":"2607ef46c8932f8baa9dc124dd9d75d893f2056e98c3ec0d6268310463f7381b"},"schema_version":"1.0"},"canonical_sha256":"c8fbe73be92dc5b7c73cee87a3c41288d3e318ad1a58f5adfb7bb7ef3a40bf78","source":{"kind":"arxiv","id":"2607.17524","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.17524","created_at":"2026-07-21T01:21:50Z"},{"alias_kind":"arxiv_version","alias_value":"2607.17524v1","created_at":"2026-07-21T01:21:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.17524","created_at":"2026-07-21T01:21:50Z"},{"alias_kind":"pith_short_12","alias_value":"ZD56OO7JFXC3","created_at":"2026-07-21T01:21:50Z"},{"alias_kind":"pith_short_16","alias_value":"ZD56OO7JFXC3PRZ4","created_at":"2026-07-21T01:21:50Z"},{"alias_kind":"pith_short_8","alias_value":"ZD56OO7J","created_at":"2026-07-21T01:21:50Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:ZD56OO7JFXC3PRZ452D2HRASRD","target":"record","payload":{"canonical_record":{"source":{"id":"2607.17524","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-20T03:57:22Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"0ba5bf7123c7a5f13182c4c58ce60065149f6ee70f7ce8705173106426b4c498","abstract_canon_sha256":"2607ef46c8932f8baa9dc124dd9d75d893f2056e98c3ec0d6268310463f7381b"},"schema_version":"1.0"},"canonical_sha256":"c8fbe73be92dc5b7c73cee87a3c41288d3e318ad1a58f5adfb7bb7ef3a40bf78","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-21T01:21:50.463782Z","signature_b64":"u/74CPchmpuWtltyrLYzcvRTZOz16PE+0Zl97Rh7LVKCld3F1tU7SNa2TF5786QomJ+XV2WyxTT65fk0mUWJCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c8fbe73be92dc5b7c73cee87a3c41288d3e318ad1a58f5adfb7bb7ef3a40bf78","last_reissued_at":"2026-07-21T01:21:50.462924Z","signature_status":"signed_v1","first_computed_at":"2026-07-21T01:21:50.462924Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.17524","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":"2026-07-21T01:21:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vHmEyMDoB5oM+7Go+IzhKV2KHqfUDx0RqXAujDEOG8zslW+EPrynALPBoZ566nFhxZoARZ+kEMHq5EzJoFNuAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T21:17:56.313743Z"},"content_sha256":"f1459885d43b55691d8390230b0f58bd5fe2067c43b24521fa01220b821c3f3a","schema_version":"1.0","event_id":"sha256:f1459885d43b55691d8390230b0f58bd5fe2067c43b24521fa01220b821c3f3a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:ZD56OO7JFXC3PRZ452D2HRASRD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Token-Level Off-Policy Learning for Faithful Generation Under Distribution Shift","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Deqing Fu, Gustavo Lucas Carvalho, Robin Jia, Zitong Huang","submitted_at":"2026-07-20T03:57:22Z","abstract_excerpt":"We propose Token-Level Off-Policy Labeling (TOPL), an off-policy training paradigm that reframes post-training as a token-level correctness prediction task. Our key intuition is that by training the model to distinguish good and bad tokens in a response, we naturally guide the model towards generating good tokens, while avoiding the pitfalls that come with directly training the model to generate off-policy tokens. Experiments on document summarization tasks show that TOPL achieves strong out-of-distribution generalization across 11 datasets against a diverse set of sequence-level and token-lev"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.17524","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/2607.17524/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":"2026-07-21T01:21:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PxWugHLzZ61HLeXgtCFJoezj4oKSEafZlFlDFp44kN8U9GT8PlTLzXKacmxlvBEA+u+c3PVZ1I5nEt9nrWkpCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T21:17:56.314352Z"},"content_sha256":"68cc5abada53553c64793b0f9765911917b6b0a06ecb5874f46f8535308f0003","schema_version":"1.0","event_id":"sha256:68cc5abada53553c64793b0f9765911917b6b0a06ecb5874f46f8535308f0003"},{"event_type":"integrity_finding","subject_pith_number":"pith:2026:ZD56OO7JFXC3PRZ452D2HRASRD","target":"integrity","payload":{"note":"Identifier '10.18653/v1/2024' is syntactically valid but the DOI registry (doi.org) returned 404, and Crossref / OpenAlex / internal corpus also have no record. The cited work could not be located through any authoritative source.","snippet":"Association for Computational Linguistics. doi: 10.18653/v1/2024. wmt-1.4. URLhttps://aclanthology.org/2024.wmt-1.4/. Paul F Christiano, Jan Leike, Tom Brown, Miljan Martic, Shane Legg, and Dario Amodei. Deep reinforcement learning from hum","arxiv_id":"2607.17524","detector":"doi_compliance","evidence":{"doi":"10.18653/v1/2024","arxiv_id":null,"ref_index":2024,"raw_excerpt":"Association for Computational Linguistics. doi: 10.18653/v1/2024. wmt-1.4. URLhttps://aclanthology.org/2024.wmt-1.4/. Paul F Christiano, Jan Leike, Tom Brown, Miljan Martic, Shane Legg, and Dario Amodei. Deep reinforcement learning from human preferences.Advances in neural information processing systems, 30,","parse_status":"well_formed","verdict_class":"cross_source","checked_sources":["crossref_by_doi","openalex_by_doi","doi_org_head"],"resolution_status":"hard_miss"},"severity":"critical","ref_index":2024,"audited_at":"2026-08-01T17:49:59.539088Z","event_type":"pith.integrity.v1","detected_doi":"10.18653/v1/2024","detector_url":"https://pith.science/pith-integrity-protocol#doi_compliance","external_url":null,"finding_type":"unresolvable_identifier","evidence_hash":"1ca2d2ebc3bcef15015bb0eee07affbe4e2c7cb2b4c39a2558888f99441e7b0c","paper_version":1,"verdict_class":"cross_source","resolved_title":null,"detector_version":"1.1.0","detected_arxiv_id":null,"integrity_event_id":16328,"payload_sha256":"81d472d8ac57ba656cdd56d2c86764276d08fcd5dbd813b3cc2dabf44de21207","signature_b64":"7uj7Grq8kyoim8nUtVG3cBt77p5yG4PJqqLaZnehn+9xEiH25BhHTVph1pH3jYnN2hkHTYaXe2Tdy+qlWDijBg==","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":"2026-08-01T17:53:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WPd0qnTOkaRxzURnrGd4cslDZkmAKQrXgxwDw6TYQUYgcQpQkCT9AymegQ5QJqC3GJ1vUpfhmZoftkrzVRLHBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T21:17:56.317183Z"},"content_sha256":"21eefe87cf19533141bcf5c7cbf7bc687dcd6eb1d9f1db7d81cc312c127158b4","schema_version":"1.0","event_id":"sha256:21eefe87cf19533141bcf5c7cbf7bc687dcd6eb1d9f1db7d81cc312c127158b4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZD56OO7JFXC3PRZ452D2HRASRD/bundle.json","state_url":"https://pith.science/pith/ZD56OO7JFXC3PRZ452D2HRASRD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZD56OO7JFXC3PRZ452D2HRASRD/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":"2026-08-03T21:17:56Z","links":{"resolver":"https://pith.science/pith/ZD56OO7JFXC3PRZ452D2HRASRD","bundle":"https://pith.science/pith/ZD56OO7JFXC3PRZ452D2HRASRD/bundle.json","state":"https://pith.science/pith/ZD56OO7JFXC3PRZ452D2HRASRD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZD56OO7JFXC3PRZ452D2HRASRD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:ZD56OO7JFXC3PRZ452D2HRASRD","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":"2607ef46c8932f8baa9dc124dd9d75d893f2056e98c3ec0d6268310463f7381b","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-20T03:57:22Z","title_canon_sha256":"0ba5bf7123c7a5f13182c4c58ce60065149f6ee70f7ce8705173106426b4c498"},"schema_version":"1.0","source":{"id":"2607.17524","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.17524","created_at":"2026-07-21T01:21:50Z"},{"alias_kind":"arxiv_version","alias_value":"2607.17524v1","created_at":"2026-07-21T01:21:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.17524","created_at":"2026-07-21T01:21:50Z"},{"alias_kind":"pith_short_12","alias_value":"ZD56OO7JFXC3","created_at":"2026-07-21T01:21:50Z"},{"alias_kind":"pith_short_16","alias_value":"ZD56OO7JFXC3PRZ4","created_at":"2026-07-21T01:21:50Z"},{"alias_kind":"pith_short_8","alias_value":"ZD56OO7J","created_at":"2026-07-21T01:21:50Z"}],"graph_snapshots":[{"event_id":"sha256:68cc5abada53553c64793b0f9765911917b6b0a06ecb5874f46f8535308f0003","target":"graph","created_at":"2026-07-21T01:21:50Z","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/2607.17524/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We propose Token-Level Off-Policy Labeling (TOPL), an off-policy training paradigm that reframes post-training as a token-level correctness prediction task. Our key intuition is that by training the model to distinguish good and bad tokens in a response, we naturally guide the model towards generating good tokens, while avoiding the pitfalls that come with directly training the model to generate off-policy tokens. Experiments on document summarization tasks show that TOPL achieves strong out-of-distribution generalization across 11 datasets against a diverse set of sequence-level and token-lev","authors_text":"Deqing Fu, Gustavo Lucas Carvalho, Robin Jia, Zitong Huang","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-20T03:57:22Z","title":"Token-Level Off-Policy Learning for Faithful Generation Under Distribution Shift"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.17524","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:f1459885d43b55691d8390230b0f58bd5fe2067c43b24521fa01220b821c3f3a","target":"record","created_at":"2026-07-21T01:21:50Z","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":"2607ef46c8932f8baa9dc124dd9d75d893f2056e98c3ec0d6268310463f7381b","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-20T03:57:22Z","title_canon_sha256":"0ba5bf7123c7a5f13182c4c58ce60065149f6ee70f7ce8705173106426b4c498"},"schema_version":"1.0","source":{"id":"2607.17524","kind":"arxiv","version":1}},"canonical_sha256":"c8fbe73be92dc5b7c73cee87a3c41288d3e318ad1a58f5adfb7bb7ef3a40bf78","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c8fbe73be92dc5b7c73cee87a3c41288d3e318ad1a58f5adfb7bb7ef3a40bf78","first_computed_at":"2026-07-21T01:21:50.462924Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-21T01:21:50.462924Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"u/74CPchmpuWtltyrLYzcvRTZOz16PE+0Zl97Rh7LVKCld3F1tU7SNa2TF5786QomJ+XV2WyxTT65fk0mUWJCA==","signature_status":"signed_v1","signed_at":"2026-07-21T01:21:50.463782Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.17524","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f1459885d43b55691d8390230b0f58bd5fe2067c43b24521fa01220b821c3f3a","sha256:68cc5abada53553c64793b0f9765911917b6b0a06ecb5874f46f8535308f0003","sha256:21eefe87cf19533141bcf5c7cbf7bc687dcd6eb1d9f1db7d81cc312c127158b4"],"state_sha256":"5ca2bcce8f0e9253dd5616a24b65638262ee8347226838484ac27fc4adad85f1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NIg0YUX0L+3LmVe9lR9fT348Gf8tjRaWK0ef6LZ4nmzCwi+/Bh3P5e6LTj+Hl+xx+mxA4oItdb8mOLp/b2bDCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T21:17:56.320039Z","bundle_sha256":"f0de9fee1b0d4f00a016cc68ae8eab9ca870787373ec19b5b7fbb9e4226d336d"}}