{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:HORC3SHC3BZHTEUKXD7B4X7IWX","short_pith_number":"pith:HORC3SHC","schema_version":"1.0","canonical_sha256":"3ba22dc8e2d87279928ab8fe1e5fe8b5cb1f710ad8c36b25560eb9e6d3543580","source":{"kind":"arxiv","id":"2409.19898","version":2},"attestation_state":"computed","paper":{"title":"UniSumEval: Towards Unified, Fine-Grained, Multi-Dimensional Summarization Evaluation for LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Hang Su, Hwanjun Song, Jason Cai, Taewon Yun, Yuho Lee","submitted_at":"2024-09-30T02:56:35Z","abstract_excerpt":"Existing benchmarks for summarization quality evaluation often lack diverse input scenarios, focus on narrowly defined dimensions (e.g., faithfulness), and struggle with subjective and coarse-grained annotation schemes. To address these shortcomings, we create UniSumEval benchmark, which extends the range of input context (e.g., domain, length) and provides fine-grained, multi-dimensional annotations. We use AI assistance in data creation, identifying potentially hallucinogenic input texts, and also helping human annotators reduce the difficulty of fine-grained annotation tasks. With UniSumEva"},"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":"2409.19898","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-09-30T02:56:35Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"7c55695fad70b033da05f35647a17fb05cdbcfb717e85debd9391259ce3e125c","abstract_canon_sha256":"68f1f5d66c70f659bd4bd23273f24776e5ae455bfbc52c8e53b1a1df9b01cf69"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:13:57.864663Z","signature_b64":"9ZOHBvKC1usC5MNLX+HNDHsrKsH9gh7/uVKazR0sE/mpXt5Gcd5t+WfLN1IOYUC8IooeyVyz5BN4g6OzNWGHDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3ba22dc8e2d87279928ab8fe1e5fe8b5cb1f710ad8c36b25560eb9e6d3543580","last_reissued_at":"2026-07-05T09:13:57.862105Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:13:57.862105Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"UniSumEval: Towards Unified, Fine-Grained, Multi-Dimensional Summarization Evaluation for LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Hang Su, Hwanjun Song, Jason Cai, Taewon Yun, Yuho Lee","submitted_at":"2024-09-30T02:56:35Z","abstract_excerpt":"Existing benchmarks for summarization quality evaluation often lack diverse input scenarios, focus on narrowly defined dimensions (e.g., faithfulness), and struggle with subjective and coarse-grained annotation schemes. To address these shortcomings, we create UniSumEval benchmark, which extends the range of input context (e.g., domain, length) and provides fine-grained, multi-dimensional annotations. We use AI assistance in data creation, identifying potentially hallucinogenic input texts, and also helping human annotators reduce the difficulty of fine-grained annotation tasks. With UniSumEva"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.19898","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/2409.19898/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":"2409.19898","created_at":"2026-07-05T09:13:57.864177+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.19898v2","created_at":"2026-07-05T09:13:57.864177+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.19898","created_at":"2026-07-05T09:13:57.864177+00:00"},{"alias_kind":"pith_short_12","alias_value":"HORC3SHC3BZH","created_at":"2026-07-05T09:13:57.864177+00:00"},{"alias_kind":"pith_short_16","alias_value":"HORC3SHC3BZHTEUK","created_at":"2026-07-05T09:13:57.864177+00:00"},{"alias_kind":"pith_short_8","alias_value":"HORC3SHC","created_at":"2026-07-05T09:13:57.864177+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.17200","citing_title":"Calibrating Model-Based Evaluation Metrics for Summarization","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17197","citing_title":"Learning to Control Summaries with Score Ranking","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HORC3SHC3BZHTEUKXD7B4X7IWX","json":"https://pith.science/pith/HORC3SHC3BZHTEUKXD7B4X7IWX.json","graph_json":"https://pith.science/api/pith-number/HORC3SHC3BZHTEUKXD7B4X7IWX/graph.json","events_json":"https://pith.science/api/pith-number/HORC3SHC3BZHTEUKXD7B4X7IWX/events.json","paper":"https://pith.science/paper/HORC3SHC"},"agent_actions":{"view_html":"https://pith.science/pith/HORC3SHC3BZHTEUKXD7B4X7IWX","download_json":"https://pith.science/pith/HORC3SHC3BZHTEUKXD7B4X7IWX.json","view_paper":"https://pith.science/paper/HORC3SHC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.19898&json=true","fetch_graph":"https://pith.science/api/pith-number/HORC3SHC3BZHTEUKXD7B4X7IWX/graph.json","fetch_events":"https://pith.science/api/pith-number/HORC3SHC3BZHTEUKXD7B4X7IWX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HORC3SHC3BZHTEUKXD7B4X7IWX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HORC3SHC3BZHTEUKXD7B4X7IWX/action/storage_attestation","attest_author":"https://pith.science/pith/HORC3SHC3BZHTEUKXD7B4X7IWX/action/author_attestation","sign_citation":"https://pith.science/pith/HORC3SHC3BZHTEUKXD7B4X7IWX/action/citation_signature","submit_replication":"https://pith.science/pith/HORC3SHC3BZHTEUKXD7B4X7IWX/action/replication_record"}},"created_at":"2026-07-05T09:13:57.864177+00:00","updated_at":"2026-07-05T09:13:57.864177+00:00"}