{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:AI5K4RANRBDKIEBHDI2FA5BROR","short_pith_number":"pith:AI5K4RAN","schema_version":"1.0","canonical_sha256":"023aae440d8846a410271a34507431747c9d7236b4a2a49a5fb74620a839cb01","source":{"kind":"arxiv","id":"1907.12894","version":1},"attestation_state":"computed","paper":{"title":"Reward Learning for Efficient Reinforcement Learning in Extractive Document Summarisation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Christian M. Meyer, Iryna Gurevych, Mohsen Mesgar, Yang Gao","submitted_at":"2019-07-30T13:31:07Z","abstract_excerpt":"Document summarisation can be formulated as a sequential decision-making problem, which can be solved by Reinforcement Learning (RL) algorithms. The predominant RL paradigm for summarisation learns a cross-input policy, which requires considerable time, data and parameter tuning due to the huge search spaces and the delayed rewards. Learning input-specific RL policies is a more efficient alternative but so far depends on handcrafted rewards, which are difficult to design and yield poor performance. We propose RELIS, a novel RL paradigm that learns a reward function with Learning-to-Rank (L2R) "},"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":"1907.12894","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-07-30T13:31:07Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"8d53ea466ab05ad9a31b31c702a88f6fb7b4d5152990bcc1c223cfe8610859ad","abstract_canon_sha256":"9f1a48e1c43e677c119331783d94edb77f3bbe2cea79d3ef94c3a662afe7b5e0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-04T23:50:25.510221Z","signature_b64":"a9yJNsRDThFoPKaKFF++3HEVGY8xkdvnhPe55ymadO/QVOESUu9A+VNeM/c8KxFxDduIKY1VyppxLe9lI/nPBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"023aae440d8846a410271a34507431747c9d7236b4a2a49a5fb74620a839cb01","last_reissued_at":"2026-07-04T23:50:25.509392Z","signature_status":"signed_v1","first_computed_at":"2026-07-04T23:50:25.509392Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reward Learning for Efficient Reinforcement Learning in Extractive Document Summarisation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Christian M. Meyer, Iryna Gurevych, Mohsen Mesgar, Yang Gao","submitted_at":"2019-07-30T13:31:07Z","abstract_excerpt":"Document summarisation can be formulated as a sequential decision-making problem, which can be solved by Reinforcement Learning (RL) algorithms. The predominant RL paradigm for summarisation learns a cross-input policy, which requires considerable time, data and parameter tuning due to the huge search spaces and the delayed rewards. Learning input-specific RL policies is a more efficient alternative but so far depends on handcrafted rewards, which are difficult to design and yield poor performance. We propose RELIS, a novel RL paradigm that learns a reward function with Learning-to-Rank (L2R) "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1907.12894","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/1907.12894/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":"1907.12894","created_at":"2026-07-04T23:50:25.509450+00:00"},{"alias_kind":"arxiv_version","alias_value":"1907.12894v1","created_at":"2026-07-04T23:50:25.509450+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1907.12894","created_at":"2026-07-04T23:50:25.509450+00:00"},{"alias_kind":"pith_short_12","alias_value":"AI5K4RANRBDK","created_at":"2026-07-04T23:50:25.509450+00:00"},{"alias_kind":"pith_short_16","alias_value":"AI5K4RANRBDKIEBH","created_at":"2026-07-04T23:50:25.509450+00:00"},{"alias_kind":"pith_short_8","alias_value":"AI5K4RAN","created_at":"2026-07-04T23:50:25.509450+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2009.01325","citing_title":"Learning to summarize from human feedback","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2409.12917","citing_title":"Training Language Models to Self-Correct via Reinforcement Learning","ref_index":84,"is_internal_anchor":false},{"citing_arxiv_id":"2309.00267","citing_title":"RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback","ref_index":81,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AI5K4RANRBDKIEBHDI2FA5BROR","json":"https://pith.science/pith/AI5K4RANRBDKIEBHDI2FA5BROR.json","graph_json":"https://pith.science/api/pith-number/AI5K4RANRBDKIEBHDI2FA5BROR/graph.json","events_json":"https://pith.science/api/pith-number/AI5K4RANRBDKIEBHDI2FA5BROR/events.json","paper":"https://pith.science/paper/AI5K4RAN"},"agent_actions":{"view_html":"https://pith.science/pith/AI5K4RANRBDKIEBHDI2FA5BROR","download_json":"https://pith.science/pith/AI5K4RANRBDKIEBHDI2FA5BROR.json","view_paper":"https://pith.science/paper/AI5K4RAN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1907.12894&json=true","fetch_graph":"https://pith.science/api/pith-number/AI5K4RANRBDKIEBHDI2FA5BROR/graph.json","fetch_events":"https://pith.science/api/pith-number/AI5K4RANRBDKIEBHDI2FA5BROR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AI5K4RANRBDKIEBHDI2FA5BROR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AI5K4RANRBDKIEBHDI2FA5BROR/action/storage_attestation","attest_author":"https://pith.science/pith/AI5K4RANRBDKIEBHDI2FA5BROR/action/author_attestation","sign_citation":"https://pith.science/pith/AI5K4RANRBDKIEBHDI2FA5BROR/action/citation_signature","submit_replication":"https://pith.science/pith/AI5K4RANRBDKIEBHDI2FA5BROR/action/replication_record"}},"created_at":"2026-07-04T23:50:25.509450+00:00","updated_at":"2026-07-04T23:50:25.509450+00:00"}