{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:DYXGPTGBOAACNE2U7KX4DJXQP3","short_pith_number":"pith:DYXGPTGB","schema_version":"1.0","canonical_sha256":"1e2e67ccc17000269354faafc1a6f07eea65da407c2c78babb389f5009983605","source":{"kind":"arxiv","id":"2309.07382","version":2},"attestation_state":"computed","paper":{"title":"Less is More for Long Document Summary Evaluation by LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Estevam Hruschka, Hayate Iso, Nikita Bhutani, Pouya Pezeshkpour, Yunshu Wu","submitted_at":"2023-09-14T01:59:15Z","abstract_excerpt":"Large Language Models (LLMs) have shown promising performance in summary evaluation tasks, yet they face challenges such as high computational costs and the Lost-in-the-Middle problem where important information in the middle of long documents is often overlooked. To address these issues, this paper introduces a novel approach, Extract-then-Evaluate, which involves extracting key sentences from a long source document and then evaluating the summary by prompting LLMs. The results reveal that the proposed method not only significantly reduces evaluation costs but also exhibits a higher correlati"},"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":"2309.07382","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-09-14T01:59:15Z","cross_cats_sorted":[],"title_canon_sha256":"21530d2ef552fcf17bd2826b0ff976956021f62d6143ec83b5fd89dc6af3fea9","abstract_canon_sha256":"b926b3b1dbe36c05e85f281948cc7516e2e8340471c65745f9252307c018fc0f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:34:50.507809Z","signature_b64":"Mhw7DkiUUCmUNtxLJHELAZlJpBX3DpKd22mM1voAPBK8pj63LPkeTMiIzcU5Arg9Dc4/ol014Kf3bQxXvtCyAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1e2e67ccc17000269354faafc1a6f07eea65da407c2c78babb389f5009983605","last_reissued_at":"2026-07-05T07:34:50.507467Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:34:50.507467Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Less is More for Long Document Summary Evaluation by LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Estevam Hruschka, Hayate Iso, Nikita Bhutani, Pouya Pezeshkpour, Yunshu Wu","submitted_at":"2023-09-14T01:59:15Z","abstract_excerpt":"Large Language Models (LLMs) have shown promising performance in summary evaluation tasks, yet they face challenges such as high computational costs and the Lost-in-the-Middle problem where important information in the middle of long documents is often overlooked. To address these issues, this paper introduces a novel approach, Extract-then-Evaluate, which involves extracting key sentences from a long source document and then evaluating the summary by prompting LLMs. The results reveal that the proposed method not only significantly reduces evaluation costs but also exhibits a higher correlati"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.07382","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/2309.07382/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":"2309.07382","created_at":"2026-07-05T07:34:50.507520+00:00"},{"alias_kind":"arxiv_version","alias_value":"2309.07382v2","created_at":"2026-07-05T07:34:50.507520+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.07382","created_at":"2026-07-05T07:34:50.507520+00:00"},{"alias_kind":"pith_short_12","alias_value":"DYXGPTGBOAAC","created_at":"2026-07-05T07:34:50.507520+00:00"},{"alias_kind":"pith_short_16","alias_value":"DYXGPTGBOAACNE2U","created_at":"2026-07-05T07:34:50.507520+00:00"},{"alias_kind":"pith_short_8","alias_value":"DYXGPTGB","created_at":"2026-07-05T07:34:50.507520+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2508.15919","citing_title":"HFX: Joint Design of Algorithms and Systems for Multi-SLO Serving and Fast Scaling","ref_index":48,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DYXGPTGBOAACNE2U7KX4DJXQP3","json":"https://pith.science/pith/DYXGPTGBOAACNE2U7KX4DJXQP3.json","graph_json":"https://pith.science/api/pith-number/DYXGPTGBOAACNE2U7KX4DJXQP3/graph.json","events_json":"https://pith.science/api/pith-number/DYXGPTGBOAACNE2U7KX4DJXQP3/events.json","paper":"https://pith.science/paper/DYXGPTGB"},"agent_actions":{"view_html":"https://pith.science/pith/DYXGPTGBOAACNE2U7KX4DJXQP3","download_json":"https://pith.science/pith/DYXGPTGBOAACNE2U7KX4DJXQP3.json","view_paper":"https://pith.science/paper/DYXGPTGB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2309.07382&json=true","fetch_graph":"https://pith.science/api/pith-number/DYXGPTGBOAACNE2U7KX4DJXQP3/graph.json","fetch_events":"https://pith.science/api/pith-number/DYXGPTGBOAACNE2U7KX4DJXQP3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DYXGPTGBOAACNE2U7KX4DJXQP3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DYXGPTGBOAACNE2U7KX4DJXQP3/action/storage_attestation","attest_author":"https://pith.science/pith/DYXGPTGBOAACNE2U7KX4DJXQP3/action/author_attestation","sign_citation":"https://pith.science/pith/DYXGPTGBOAACNE2U7KX4DJXQP3/action/citation_signature","submit_replication":"https://pith.science/pith/DYXGPTGBOAACNE2U7KX4DJXQP3/action/replication_record"}},"created_at":"2026-07-05T07:34:50.507520+00:00","updated_at":"2026-07-05T07:34:50.507520+00:00"}