{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:7XCKTO7T2C3QDI3KMHKCKGMH2Z","short_pith_number":"pith:7XCKTO7T","schema_version":"1.0","canonical_sha256":"fdc4a9bbf3d0b701a36a61d4251987d65318dd115ca96cc3a1cf4a39ce91f2ee","source":{"kind":"arxiv","id":"2408.04237","version":2},"attestation_state":"computed","paper":{"title":"Learning to Rewrite: Generalized LLM-Generated Text Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chengzhi Mao, Junfeng Yang, Ran Li, Wei Hao, Weiliang Zhao","submitted_at":"2024-08-08T05:53:39Z","abstract_excerpt":"Large language models (LLMs) present significant risks when used to generate non-factual content and spread disinformation at scale. Detecting such LLM-generated content is crucial, yet current detectors often struggle to generalize in open-world contexts. We introduce Learning2Rewrite, a novel framework for detecting AI-generated text with exceptional generalization to unseen domains. Our method leverages the insight that LLMs inherently modify AI-generated content less than human-written text when tasked with rewriting. By training LLMs to minimize alterations on AI-generated inputs, we ampl"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2408.04237","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-08-08T05:53:39Z","cross_cats_sorted":[],"title_canon_sha256":"fb1649cfea5d5111081a8d13ab5e569924dc6d73df35b76667a013d59f8a9e45","abstract_canon_sha256":"90daab020f1fefdbe060aaf00275d277d4328e1e268d160ca99f61d2a37c13a1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:14:44.127093Z","signature_b64":"bP7m1ChDCushg2QQpq0QTcdJG8opS6w1bjXyBbLCXkTxiTl1YkTB9zyu/pldADnHoiCi+Srh1Mpw/Xy0DUO9Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fdc4a9bbf3d0b701a36a61d4251987d65318dd115ca96cc3a1cf4a39ce91f2ee","last_reissued_at":"2026-07-05T10:14:44.126528Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:14:44.126528Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning to Rewrite: Generalized LLM-Generated Text Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chengzhi Mao, Junfeng Yang, Ran Li, Wei Hao, Weiliang Zhao","submitted_at":"2024-08-08T05:53:39Z","abstract_excerpt":"Large language models (LLMs) present significant risks when used to generate non-factual content and spread disinformation at scale. Detecting such LLM-generated content is crucial, yet current detectors often struggle to generalize in open-world contexts. We introduce Learning2Rewrite, a novel framework for detecting AI-generated text with exceptional generalization to unseen domains. Our method leverages the insight that LLMs inherently modify AI-generated content less than human-written text when tasked with rewriting. By training LLMs to minimize alterations on AI-generated inputs, we ampl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.04237","kind":"arxiv","version":2},"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/2408.04237/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2408.04237","created_at":"2026-07-05T10:14:44.126584+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.04237v2","created_at":"2026-07-05T10:14:44.126584+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.04237","created_at":"2026-07-05T10:14:44.126584+00:00"},{"alias_kind":"pith_short_12","alias_value":"7XCKTO7T2C3Q","created_at":"2026-07-05T10:14:44.126584+00:00"},{"alias_kind":"pith_short_16","alias_value":"7XCKTO7T2C3QDI3K","created_at":"2026-07-05T10:14:44.126584+00:00"},{"alias_kind":"pith_short_8","alias_value":"7XCKTO7T","created_at":"2026-07-05T10:14:44.126584+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.04205","citing_title":"DetectZoo: A Unified Toolkit for AI-Generated Content Detection Across Text, Audio, and Image Modalities","ref_index":82,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7XCKTO7T2C3QDI3KMHKCKGMH2Z","json":"https://pith.science/pith/7XCKTO7T2C3QDI3KMHKCKGMH2Z.json","graph_json":"https://pith.science/api/pith-number/7XCKTO7T2C3QDI3KMHKCKGMH2Z/graph.json","events_json":"https://pith.science/api/pith-number/7XCKTO7T2C3QDI3KMHKCKGMH2Z/events.json","paper":"https://pith.science/paper/7XCKTO7T"},"agent_actions":{"view_html":"https://pith.science/pith/7XCKTO7T2C3QDI3KMHKCKGMH2Z","download_json":"https://pith.science/pith/7XCKTO7T2C3QDI3KMHKCKGMH2Z.json","view_paper":"https://pith.science/paper/7XCKTO7T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.04237&json=true","fetch_graph":"https://pith.science/api/pith-number/7XCKTO7T2C3QDI3KMHKCKGMH2Z/graph.json","fetch_events":"https://pith.science/api/pith-number/7XCKTO7T2C3QDI3KMHKCKGMH2Z/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7XCKTO7T2C3QDI3KMHKCKGMH2Z/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7XCKTO7T2C3QDI3KMHKCKGMH2Z/action/storage_attestation","attest_author":"https://pith.science/pith/7XCKTO7T2C3QDI3KMHKCKGMH2Z/action/author_attestation","sign_citation":"https://pith.science/pith/7XCKTO7T2C3QDI3KMHKCKGMH2Z/action/citation_signature","submit_replication":"https://pith.science/pith/7XCKTO7T2C3QDI3KMHKCKGMH2Z/action/replication_record"}},"created_at":"2026-07-05T10:14:44.126584+00:00","updated_at":"2026-07-05T10:14:44.126584+00:00"}