{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:5LLLNXINMQBL2HY6JU7RI7IAF4","short_pith_number":"pith:5LLLNXIN","schema_version":"1.0","canonical_sha256":"ead6b6dd0d6402bd1f1e4d3f147d002f32869c8f179cdf7e47d9c5578d504833","source":{"kind":"arxiv","id":"1808.10792","version":2},"attestation_state":"computed","paper":{"title":"Bottom-Up Abstractive Summarization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Alexander M. Rush, Sebastian Gehrmann, Yuntian Deng","submitted_at":"2018-08-31T14:55:52Z","abstract_excerpt":"Neural network-based methods for abstractive summarization produce outputs that are more fluent than other techniques, but which can be poor at content selection. This work proposes a simple technique for addressing this issue: use a data-efficient content selector to over-determine phrases in a source document that should be part of the summary. We use this selector as a bottom-up attention step to constrain the model to likely phrases. We show that this approach improves the ability to compress text, while still generating fluent summaries. This two-step process is both simpler and higher pe"},"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":"1808.10792","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-08-31T14:55:52Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"6e1c90d18daff81173fee684cbc2fb359bbfc6876d6e3cfc86df8c5da0848985","abstract_canon_sha256":"96c43f1218f1b83ae3c1910c0e10c32837be31deaa5c01a87fbeead3f97f33a4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:03:45.335730Z","signature_b64":"9/w61FqREnsmjvD727WlDj5WdXv1K36MwaGG1Y80oFt03oHOZhyPvJBSg2wsFsCQVUnJSo78NcR8h2sOiuFSDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ead6b6dd0d6402bd1f1e4d3f147d002f32869c8f179cdf7e47d9c5578d504833","last_reissued_at":"2026-05-18T00:03:45.335231Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:03:45.335231Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bottom-Up Abstractive Summarization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Alexander M. Rush, Sebastian Gehrmann, Yuntian Deng","submitted_at":"2018-08-31T14:55:52Z","abstract_excerpt":"Neural network-based methods for abstractive summarization produce outputs that are more fluent than other techniques, but which can be poor at content selection. This work proposes a simple technique for addressing this issue: use a data-efficient content selector to over-determine phrases in a source document that should be part of the summary. We use this selector as a bottom-up attention step to constrain the model to likely phrases. We show that this approach improves the ability to compress text, while still generating fluent summaries. This two-step process is both simpler and higher pe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1808.10792","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":""},"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":"1808.10792","created_at":"2026-05-18T00:03:45.335309+00:00"},{"alias_kind":"arxiv_version","alias_value":"1808.10792v2","created_at":"2026-05-18T00:03:45.335309+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1808.10792","created_at":"2026-05-18T00:03:45.335309+00:00"},{"alias_kind":"pith_short_12","alias_value":"5LLLNXINMQBL","created_at":"2026-05-18T12:32:08.215937+00:00"},{"alias_kind":"pith_short_16","alias_value":"5LLLNXINMQBL2HY6","created_at":"2026-05-18T12:32:08.215937+00:00"},{"alias_kind":"pith_short_8","alias_value":"5LLLNXIN","created_at":"2026-05-18T12:32:08.215937+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2109.10616","citing_title":"Enriching and Controlling Global Semantics for Text Summarization","ref_index":12,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5LLLNXINMQBL2HY6JU7RI7IAF4","json":"https://pith.science/pith/5LLLNXINMQBL2HY6JU7RI7IAF4.json","graph_json":"https://pith.science/api/pith-number/5LLLNXINMQBL2HY6JU7RI7IAF4/graph.json","events_json":"https://pith.science/api/pith-number/5LLLNXINMQBL2HY6JU7RI7IAF4/events.json","paper":"https://pith.science/paper/5LLLNXIN"},"agent_actions":{"view_html":"https://pith.science/pith/5LLLNXINMQBL2HY6JU7RI7IAF4","download_json":"https://pith.science/pith/5LLLNXINMQBL2HY6JU7RI7IAF4.json","view_paper":"https://pith.science/paper/5LLLNXIN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1808.10792&json=true","fetch_graph":"https://pith.science/api/pith-number/5LLLNXINMQBL2HY6JU7RI7IAF4/graph.json","fetch_events":"https://pith.science/api/pith-number/5LLLNXINMQBL2HY6JU7RI7IAF4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5LLLNXINMQBL2HY6JU7RI7IAF4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5LLLNXINMQBL2HY6JU7RI7IAF4/action/storage_attestation","attest_author":"https://pith.science/pith/5LLLNXINMQBL2HY6JU7RI7IAF4/action/author_attestation","sign_citation":"https://pith.science/pith/5LLLNXINMQBL2HY6JU7RI7IAF4/action/citation_signature","submit_replication":"https://pith.science/pith/5LLLNXINMQBL2HY6JU7RI7IAF4/action/replication_record"}},"created_at":"2026-05-18T00:03:45.335309+00:00","updated_at":"2026-05-18T00:03:45.335309+00:00"}