{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:ASPLGO6INTQFHFSXDFT2WA774B","short_pith_number":"pith:ASPLGO6I","schema_version":"1.0","canonical_sha256":"049eb33bc86ce05396571967ab03ffe073ba10dee907d52afbe2f3baa6b8b16a","source":{"kind":"arxiv","id":"2108.01850","version":1},"attestation_state":"computed","paper":{"title":"Controlled Text Generation as Continuous Optimization with Multiple Constraints","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Aliaksei Severyn, Eric Malmi, Sachin Kumar, Yulia Tsvetkov","submitted_at":"2021-08-04T05:25:20Z","abstract_excerpt":"As large-scale language model pretraining pushes the state-of-the-art in text generation, recent work has turned to controlling attributes of the text such models generate. While modifying the pretrained models via fine-tuning remains the popular approach, it incurs a significant computational cost and can be infeasible due to lack of appropriate data. As an alternative, we propose MuCoCO -- a flexible and modular algorithm for controllable inference from pretrained models. We formulate the decoding process as an optimization problem which allows for multiple attributes we aim to control to be"},"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":"2108.01850","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-08-04T05:25:20Z","cross_cats_sorted":[],"title_canon_sha256":"160687f7ad4caaa57f90107d1bec122f12363877c5dd8ee9ea154d651abec8e0","abstract_canon_sha256":"53bb9bb6b81d50e841453c68a9e87b1e2c6e1037b4574fda05754268d569f0ea"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:03:11.455360Z","signature_b64":"vC5OTCKdVJ9flDFSwHQERJVjW7KgfPvZnWXw19EMeoCrpV0V+lCFEMyXSnR3ErXWsi2BZIYjYJ+jh32jtgDzDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"049eb33bc86ce05396571967ab03ffe073ba10dee907d52afbe2f3baa6b8b16a","last_reissued_at":"2026-07-05T03:03:11.454960Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:03:11.454960Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Controlled Text Generation as Continuous Optimization with Multiple Constraints","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Aliaksei Severyn, Eric Malmi, Sachin Kumar, Yulia Tsvetkov","submitted_at":"2021-08-04T05:25:20Z","abstract_excerpt":"As large-scale language model pretraining pushes the state-of-the-art in text generation, recent work has turned to controlling attributes of the text such models generate. While modifying the pretrained models via fine-tuning remains the popular approach, it incurs a significant computational cost and can be infeasible due to lack of appropriate data. As an alternative, we propose MuCoCO -- a flexible and modular algorithm for controllable inference from pretrained models. We formulate the decoding process as an optimization problem which allows for multiple attributes we aim to control to be"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2108.01850","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/2108.01850/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":"2108.01850","created_at":"2026-07-05T03:03:11.455024+00:00"},{"alias_kind":"arxiv_version","alias_value":"2108.01850v1","created_at":"2026-07-05T03:03:11.455024+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2108.01850","created_at":"2026-07-05T03:03:11.455024+00:00"},{"alias_kind":"pith_short_12","alias_value":"ASPLGO6INTQF","created_at":"2026-07-05T03:03:11.455024+00:00"},{"alias_kind":"pith_short_16","alias_value":"ASPLGO6INTQFHFSX","created_at":"2026-07-05T03:03:11.455024+00:00"},{"alias_kind":"pith_short_8","alias_value":"ASPLGO6I","created_at":"2026-07-05T03:03:11.455024+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.08071","citing_title":"COBART: Controlled, Optimized, Bidirectional and Auto-Regressive Transformer for Ad Headline Generation","ref_index":16,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ASPLGO6INTQFHFSXDFT2WA774B","json":"https://pith.science/pith/ASPLGO6INTQFHFSXDFT2WA774B.json","graph_json":"https://pith.science/api/pith-number/ASPLGO6INTQFHFSXDFT2WA774B/graph.json","events_json":"https://pith.science/api/pith-number/ASPLGO6INTQFHFSXDFT2WA774B/events.json","paper":"https://pith.science/paper/ASPLGO6I"},"agent_actions":{"view_html":"https://pith.science/pith/ASPLGO6INTQFHFSXDFT2WA774B","download_json":"https://pith.science/pith/ASPLGO6INTQFHFSXDFT2WA774B.json","view_paper":"https://pith.science/paper/ASPLGO6I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2108.01850&json=true","fetch_graph":"https://pith.science/api/pith-number/ASPLGO6INTQFHFSXDFT2WA774B/graph.json","fetch_events":"https://pith.science/api/pith-number/ASPLGO6INTQFHFSXDFT2WA774B/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ASPLGO6INTQFHFSXDFT2WA774B/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ASPLGO6INTQFHFSXDFT2WA774B/action/storage_attestation","attest_author":"https://pith.science/pith/ASPLGO6INTQFHFSXDFT2WA774B/action/author_attestation","sign_citation":"https://pith.science/pith/ASPLGO6INTQFHFSXDFT2WA774B/action/citation_signature","submit_replication":"https://pith.science/pith/ASPLGO6INTQFHFSXDFT2WA774B/action/replication_record"}},"created_at":"2026-07-05T03:03:11.455024+00:00","updated_at":"2026-07-05T03:03:11.455024+00:00"}