{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:STODHTOKKZKJCRZXLMPHB7XJFL","short_pith_number":"pith:STODHTOK","schema_version":"1.0","canonical_sha256":"94dc33cdca56549147375b1e70fee92addf080373528633d14b14adf36ffafbc","source":{"kind":"arxiv","id":"2204.05512","version":2},"attestation_state":"computed","paper":{"title":"Make The Most of Prior Data: A Solution for Interactive Text Summarization with Preference Feedback","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bao-Sinh Nguyen, Dung Tien Le, Duy-Hung Nguyen, Hung Le, Minh-Tien Nguyen, Nguyen Viet Dung Nghiem, Shahab Sabahi","submitted_at":"2022-04-12T03:56:59Z","abstract_excerpt":"For summarization, human preference is critical to tame outputs of the summarizer in favor of human interests, as ground-truth summaries are scarce and ambiguous. Practical settings require dynamic exchanges between human and AI agent wherein feedback is provided in an online manner, a few at a time. In this paper, we introduce a new framework to train summarization models with preference feedback interactively. By properly leveraging offline data and a novel reward model, we improve the performance regarding ROUGE scores and sample-efficiency. Our experiments on three various datasets confirm"},"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":"2204.05512","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2022-04-12T03:56:59Z","cross_cats_sorted":[],"title_canon_sha256":"4ee323a4f3c275d116c96accb44f71dd925414325f564df41ebbd758c452a54c","abstract_canon_sha256":"396fe70c31871f8005f0f51b2f2ee89e1490959b33b0137912bf39417f1fbc2c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:22:41.050065Z","signature_b64":"NhlGipH3oxQnp+BQpPira4s2zuhy7uZ49IZTb2AzHC5h7hGIo4+P5kw5Oa28SX85kMj4Ft4N4SB5JlCB/eq4CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"94dc33cdca56549147375b1e70fee92addf080373528633d14b14adf36ffafbc","last_reissued_at":"2026-07-05T04:22:41.049630Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:22:41.049630Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Make The Most of Prior Data: A Solution for Interactive Text Summarization with Preference Feedback","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bao-Sinh Nguyen, Dung Tien Le, Duy-Hung Nguyen, Hung Le, Minh-Tien Nguyen, Nguyen Viet Dung Nghiem, Shahab Sabahi","submitted_at":"2022-04-12T03:56:59Z","abstract_excerpt":"For summarization, human preference is critical to tame outputs of the summarizer in favor of human interests, as ground-truth summaries are scarce and ambiguous. Practical settings require dynamic exchanges between human and AI agent wherein feedback is provided in an online manner, a few at a time. In this paper, we introduce a new framework to train summarization models with preference feedback interactively. By properly leveraging offline data and a novel reward model, we improve the performance regarding ROUGE scores and sample-efficiency. Our experiments on three various datasets confirm"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2204.05512","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/2204.05512/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":"2204.05512","created_at":"2026-07-05T04:22:41.049686+00:00"},{"alias_kind":"arxiv_version","alias_value":"2204.05512v2","created_at":"2026-07-05T04:22:41.049686+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2204.05512","created_at":"2026-07-05T04:22:41.049686+00:00"},{"alias_kind":"pith_short_12","alias_value":"STODHTOKKZKJ","created_at":"2026-07-05T04:22:41.049686+00:00"},{"alias_kind":"pith_short_16","alias_value":"STODHTOKKZKJCRZX","created_at":"2026-07-05T04:22:41.049686+00:00"},{"alias_kind":"pith_short_8","alias_value":"STODHTOK","created_at":"2026-07-05T04:22:41.049686+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/STODHTOKKZKJCRZXLMPHB7XJFL","json":"https://pith.science/pith/STODHTOKKZKJCRZXLMPHB7XJFL.json","graph_json":"https://pith.science/api/pith-number/STODHTOKKZKJCRZXLMPHB7XJFL/graph.json","events_json":"https://pith.science/api/pith-number/STODHTOKKZKJCRZXLMPHB7XJFL/events.json","paper":"https://pith.science/paper/STODHTOK"},"agent_actions":{"view_html":"https://pith.science/pith/STODHTOKKZKJCRZXLMPHB7XJFL","download_json":"https://pith.science/pith/STODHTOKKZKJCRZXLMPHB7XJFL.json","view_paper":"https://pith.science/paper/STODHTOK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2204.05512&json=true","fetch_graph":"https://pith.science/api/pith-number/STODHTOKKZKJCRZXLMPHB7XJFL/graph.json","fetch_events":"https://pith.science/api/pith-number/STODHTOKKZKJCRZXLMPHB7XJFL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/STODHTOKKZKJCRZXLMPHB7XJFL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/STODHTOKKZKJCRZXLMPHB7XJFL/action/storage_attestation","attest_author":"https://pith.science/pith/STODHTOKKZKJCRZXLMPHB7XJFL/action/author_attestation","sign_citation":"https://pith.science/pith/STODHTOKKZKJCRZXLMPHB7XJFL/action/citation_signature","submit_replication":"https://pith.science/pith/STODHTOKKZKJCRZXLMPHB7XJFL/action/replication_record"}},"created_at":"2026-07-05T04:22:41.049686+00:00","updated_at":"2026-07-05T04:22:41.049686+00:00"}