{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:35T2QTCO7LB4CYWXBGKQ2ENBJS","short_pith_number":"pith:35T2QTCO","schema_version":"1.0","canonical_sha256":"df67a84c4efac3c162d709950d11a14cadec528c03b2bef02800998b96bf057c","source":{"kind":"arxiv","id":"2211.07343","version":2},"attestation_state":"computed","paper":{"title":"Replacing Language Model for Style Transfer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Pengyu Cheng, Ruineng Li","submitted_at":"2022-11-14T13:35:55Z","abstract_excerpt":"We introduce replacing language model (RLM), a sequence-to-sequence language modeling framework for text style transfer (TST). Our method autoregressively replaces each token of the source sentence with a text span that has a similar meaning but in the target style. The new span is generated via a non-autoregressive masked language model, which can better preserve the local-contextual meaning of the replaced token. This RLM generation scheme gathers the flexibility of autoregressive models and the accuracy of non-autoregressive models, which bridges the gap between sentence-level and word-leve"},"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":"2211.07343","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-11-14T13:35:55Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"93c758babd1b35005f9e1ccd5929d8f7fe48189ae86fbf0eb28e01cf79b785f6","abstract_canon_sha256":"0f6fcde346d2bf9d88703d47f78b52e75b97c58ede96213de0f0d3ad97ff82b2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:49:59.361099Z","signature_b64":"v+8MqMJF+KYKVyAamkt3RRBkFX/jTSxKAH8Gips59CR8Lhn6zo/ZZm4nFGy/sl19r9+VvkhOF1zPh2X/78Z0Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"df67a84c4efac3c162d709950d11a14cadec528c03b2bef02800998b96bf057c","last_reissued_at":"2026-07-05T07:49:59.360659Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:49:59.360659Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Replacing Language Model for Style Transfer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Pengyu Cheng, Ruineng Li","submitted_at":"2022-11-14T13:35:55Z","abstract_excerpt":"We introduce replacing language model (RLM), a sequence-to-sequence language modeling framework for text style transfer (TST). Our method autoregressively replaces each token of the source sentence with a text span that has a similar meaning but in the target style. The new span is generated via a non-autoregressive masked language model, which can better preserve the local-contextual meaning of the replaced token. This RLM generation scheme gathers the flexibility of autoregressive models and the accuracy of non-autoregressive models, which bridges the gap between sentence-level and word-leve"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.07343","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/2211.07343/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":"2211.07343","created_at":"2026-07-05T07:49:59.360716+00:00"},{"alias_kind":"arxiv_version","alias_value":"2211.07343v2","created_at":"2026-07-05T07:49:59.360716+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.07343","created_at":"2026-07-05T07:49:59.360716+00:00"},{"alias_kind":"pith_short_12","alias_value":"35T2QTCO7LB4","created_at":"2026-07-05T07:49:59.360716+00:00"},{"alias_kind":"pith_short_16","alias_value":"35T2QTCO7LB4CYWX","created_at":"2026-07-05T07:49:59.360716+00:00"},{"alias_kind":"pith_short_8","alias_value":"35T2QTCO","created_at":"2026-07-05T07:49:59.360716+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.09329","citing_title":"Towards Efficient and Effective Alignment of Large Language Models","ref_index":29,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/35T2QTCO7LB4CYWXBGKQ2ENBJS","json":"https://pith.science/pith/35T2QTCO7LB4CYWXBGKQ2ENBJS.json","graph_json":"https://pith.science/api/pith-number/35T2QTCO7LB4CYWXBGKQ2ENBJS/graph.json","events_json":"https://pith.science/api/pith-number/35T2QTCO7LB4CYWXBGKQ2ENBJS/events.json","paper":"https://pith.science/paper/35T2QTCO"},"agent_actions":{"view_html":"https://pith.science/pith/35T2QTCO7LB4CYWXBGKQ2ENBJS","download_json":"https://pith.science/pith/35T2QTCO7LB4CYWXBGKQ2ENBJS.json","view_paper":"https://pith.science/paper/35T2QTCO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2211.07343&json=true","fetch_graph":"https://pith.science/api/pith-number/35T2QTCO7LB4CYWXBGKQ2ENBJS/graph.json","fetch_events":"https://pith.science/api/pith-number/35T2QTCO7LB4CYWXBGKQ2ENBJS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/35T2QTCO7LB4CYWXBGKQ2ENBJS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/35T2QTCO7LB4CYWXBGKQ2ENBJS/action/storage_attestation","attest_author":"https://pith.science/pith/35T2QTCO7LB4CYWXBGKQ2ENBJS/action/author_attestation","sign_citation":"https://pith.science/pith/35T2QTCO7LB4CYWXBGKQ2ENBJS/action/citation_signature","submit_replication":"https://pith.science/pith/35T2QTCO7LB4CYWXBGKQ2ENBJS/action/replication_record"}},"created_at":"2026-07-05T07:49:59.360716+00:00","updated_at":"2026-07-05T07:49:59.360716+00:00"}