{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:5AKL4ALWDUCYZMJUMSXEVMQHM5","short_pith_number":"pith:5AKL4ALW","schema_version":"1.0","canonical_sha256":"e814be01761d058cb13464ae4ab20767457463d44676d974175ba84cd4e2e0f8","source":{"kind":"arxiv","id":"2403.00827","version":1},"attestation_state":"computed","paper":{"title":"Self-Refinement of Language Models from External Proxy Metrics Feedback","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Asim Munawar, Keshav Ramji, Md Arafat Sultan, Radu Florian, Ram\\'on Fernandez Astudillo, Salim Roukos, Tahira Naseem, Young-Suk Lee","submitted_at":"2024-02-27T19:13:01Z","abstract_excerpt":"It is often desirable for Large Language Models (LLMs) to capture multiple objectives when providing a response. In document-grounded response generation, for example, agent responses are expected to be relevant to a user's query while also being grounded in a given document. In this paper, we introduce Proxy Metric-based Self-Refinement (ProMiSe), which enables an LLM to refine its own initial response along key dimensions of quality guided by external metrics feedback, yielding an overall better final response. ProMiSe leverages feedback on response quality through principle-specific proxy m"},"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":"2403.00827","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-27T19:13:01Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"d31b8e29509bb367186812dd3743c92ed707c8a0345d850af55057468d4522ff","abstract_canon_sha256":"7eae2896a7faa5de211b4c8b320a6d2e4e9a4829780fa1cc08db8aa95a0fc7f4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:51:12.315719Z","signature_b64":"M+F3x10GMVWXeGGzvMJ18iwotCANxDLZ5SXnh8y82zBl/+TkgLYam9HoO3ps1zLMI5cbBEipnvbpCsPIPlGrCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e814be01761d058cb13464ae4ab20767457463d44676d974175ba84cd4e2e0f8","last_reissued_at":"2026-07-05T07:51:12.315251Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:51:12.315251Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Self-Refinement of Language Models from External Proxy Metrics Feedback","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Asim Munawar, Keshav Ramji, Md Arafat Sultan, Radu Florian, Ram\\'on Fernandez Astudillo, Salim Roukos, Tahira Naseem, Young-Suk Lee","submitted_at":"2024-02-27T19:13:01Z","abstract_excerpt":"It is often desirable for Large Language Models (LLMs) to capture multiple objectives when providing a response. In document-grounded response generation, for example, agent responses are expected to be relevant to a user's query while also being grounded in a given document. In this paper, we introduce Proxy Metric-based Self-Refinement (ProMiSe), which enables an LLM to refine its own initial response along key dimensions of quality guided by external metrics feedback, yielding an overall better final response. ProMiSe leverages feedback on response quality through principle-specific proxy m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.00827","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/2403.00827/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":"2403.00827","created_at":"2026-07-05T07:51:12.315311+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.00827v1","created_at":"2026-07-05T07:51:12.315311+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.00827","created_at":"2026-07-05T07:51:12.315311+00:00"},{"alias_kind":"pith_short_12","alias_value":"5AKL4ALWDUCY","created_at":"2026-07-05T07:51:12.315311+00:00"},{"alias_kind":"pith_short_16","alias_value":"5AKL4ALWDUCYZMJU","created_at":"2026-07-05T07:51:12.315311+00:00"},{"alias_kind":"pith_short_8","alias_value":"5AKL4ALW","created_at":"2026-07-05T07:51:12.315311+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5AKL4ALWDUCYZMJUMSXEVMQHM5","json":"https://pith.science/pith/5AKL4ALWDUCYZMJUMSXEVMQHM5.json","graph_json":"https://pith.science/api/pith-number/5AKL4ALWDUCYZMJUMSXEVMQHM5/graph.json","events_json":"https://pith.science/api/pith-number/5AKL4ALWDUCYZMJUMSXEVMQHM5/events.json","paper":"https://pith.science/paper/5AKL4ALW"},"agent_actions":{"view_html":"https://pith.science/pith/5AKL4ALWDUCYZMJUMSXEVMQHM5","download_json":"https://pith.science/pith/5AKL4ALWDUCYZMJUMSXEVMQHM5.json","view_paper":"https://pith.science/paper/5AKL4ALW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.00827&json=true","fetch_graph":"https://pith.science/api/pith-number/5AKL4ALWDUCYZMJUMSXEVMQHM5/graph.json","fetch_events":"https://pith.science/api/pith-number/5AKL4ALWDUCYZMJUMSXEVMQHM5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5AKL4ALWDUCYZMJUMSXEVMQHM5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5AKL4ALWDUCYZMJUMSXEVMQHM5/action/storage_attestation","attest_author":"https://pith.science/pith/5AKL4ALWDUCYZMJUMSXEVMQHM5/action/author_attestation","sign_citation":"https://pith.science/pith/5AKL4ALWDUCYZMJUMSXEVMQHM5/action/citation_signature","submit_replication":"https://pith.science/pith/5AKL4ALWDUCYZMJUMSXEVMQHM5/action/replication_record"}},"created_at":"2026-07-05T07:51:12.315311+00:00","updated_at":"2026-07-05T07:51:12.315311+00:00"}