{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:3LIV44L2E6DE3WGVEXGAOODBLK","short_pith_number":"pith:3LIV44L2","schema_version":"1.0","canonical_sha256":"dad15e717a27864dd8d525cc0738615abadd6f1b66cf30bcf6bc6293093ced7b","source":{"kind":"arxiv","id":"2309.09558","version":1},"attestation_state":"computed","paper":{"title":"Summarization is (Almost) Dead","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Mingqi Gao, Xiaojun Wan, Xiao Pu","submitted_at":"2023-09-18T08:13:01Z","abstract_excerpt":"How well can large language models (LLMs) generate summaries? We develop new datasets and conduct human evaluation experiments to evaluate the zero-shot generation capability of LLMs across five distinct summarization tasks. Our findings indicate a clear preference among human evaluators for LLM-generated summaries over human-written summaries and summaries generated by fine-tuned models. Specifically, LLM-generated summaries exhibit better factual consistency and fewer instances of extrinsic hallucinations. Due to the satisfactory performance of LLMs in summarization tasks (even surpassing th"},"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":"2309.09558","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-09-18T08:13:01Z","cross_cats_sorted":[],"title_canon_sha256":"a9bdced1a46a0631aba7435b90a8528754b4625963b4bea8d10470effa3888e8","abstract_canon_sha256":"3dabdf7f7e01cd085c8b224f97be16290c559c0ec8db97922314c89b3da26e1c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:51:41.060151Z","signature_b64":"lWRYYRukc7+cIxaqCjmpcPJhaeRrmwVcwt8YiRLrBPcDxSgl9nhzRBYoA/qBqttmX1nYkIFLl4HlNrz2wx8eBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dad15e717a27864dd8d525cc0738615abadd6f1b66cf30bcf6bc6293093ced7b","last_reissued_at":"2026-07-05T06:51:41.059759Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:51:41.059759Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Summarization is (Almost) Dead","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Mingqi Gao, Xiaojun Wan, Xiao Pu","submitted_at":"2023-09-18T08:13:01Z","abstract_excerpt":"How well can large language models (LLMs) generate summaries? We develop new datasets and conduct human evaluation experiments to evaluate the zero-shot generation capability of LLMs across five distinct summarization tasks. Our findings indicate a clear preference among human evaluators for LLM-generated summaries over human-written summaries and summaries generated by fine-tuned models. Specifically, LLM-generated summaries exhibit better factual consistency and fewer instances of extrinsic hallucinations. Due to the satisfactory performance of LLMs in summarization tasks (even surpassing th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.09558","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/2309.09558/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":"2309.09558","created_at":"2026-07-05T06:51:41.059818+00:00"},{"alias_kind":"arxiv_version","alias_value":"2309.09558v1","created_at":"2026-07-05T06:51:41.059818+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.09558","created_at":"2026-07-05T06:51:41.059818+00:00"},{"alias_kind":"pith_short_12","alias_value":"3LIV44L2E6DE","created_at":"2026-07-05T06:51:41.059818+00:00"},{"alias_kind":"pith_short_16","alias_value":"3LIV44L2E6DE3WGV","created_at":"2026-07-05T06:51:41.059818+00:00"},{"alias_kind":"pith_short_8","alias_value":"3LIV44L2","created_at":"2026-07-05T06:51:41.059818+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.08000","citing_title":"Summarization is Not Dead Yet","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01252","citing_title":"Understanding LLM Behavior in Multi-Target Cross-Lingual Summarization","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2406.15809","citing_title":"LaMSUM: Amplifying Voices Against Harassment through LLM Guided Extractive Summarization of User Incident Reports","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17652","citing_title":"Beyond Transcripts: Iterative Peer-Editing with Audio Unlocks High-Quality Human Summaries of Conversational Speech","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02596","citing_title":"An Empirical Study of Many-Shot In-Context Learning for Machine Translation of Low-Resource Languages","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12108","citing_title":"LLM-Based Automated Diagnosis Of Integration Test Failures At Google","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20131","citing_title":"Whose Story Gets Told? Positionality and Bias in LLM Summaries of Life Narratives","ref_index":60,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3LIV44L2E6DE3WGVEXGAOODBLK","json":"https://pith.science/pith/3LIV44L2E6DE3WGVEXGAOODBLK.json","graph_json":"https://pith.science/api/pith-number/3LIV44L2E6DE3WGVEXGAOODBLK/graph.json","events_json":"https://pith.science/api/pith-number/3LIV44L2E6DE3WGVEXGAOODBLK/events.json","paper":"https://pith.science/paper/3LIV44L2"},"agent_actions":{"view_html":"https://pith.science/pith/3LIV44L2E6DE3WGVEXGAOODBLK","download_json":"https://pith.science/pith/3LIV44L2E6DE3WGVEXGAOODBLK.json","view_paper":"https://pith.science/paper/3LIV44L2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2309.09558&json=true","fetch_graph":"https://pith.science/api/pith-number/3LIV44L2E6DE3WGVEXGAOODBLK/graph.json","fetch_events":"https://pith.science/api/pith-number/3LIV44L2E6DE3WGVEXGAOODBLK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3LIV44L2E6DE3WGVEXGAOODBLK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3LIV44L2E6DE3WGVEXGAOODBLK/action/storage_attestation","attest_author":"https://pith.science/pith/3LIV44L2E6DE3WGVEXGAOODBLK/action/author_attestation","sign_citation":"https://pith.science/pith/3LIV44L2E6DE3WGVEXGAOODBLK/action/citation_signature","submit_replication":"https://pith.science/pith/3LIV44L2E6DE3WGVEXGAOODBLK/action/replication_record"}},"created_at":"2026-07-05T06:51:41.059818+00:00","updated_at":"2026-07-05T06:51:41.059818+00:00"}