{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:XSMADUCB6KDQDJLU5HZ3ZZMRX7","short_pith_number":"pith:XSMADUCB","schema_version":"1.0","canonical_sha256":"bc9801d041f28701a574e9f3bce591bffa1a9802383c425e10c3040738f0651a","source":{"kind":"arxiv","id":"2409.02076","version":7},"attestation_state":"computed","paper":{"title":"LongGenBench: Benchmarking Long-Form Generation in Long Context LLMs","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ming Shan Hee, Roy Ka-Wei Lee, Yuhao Wu, Zhiqing Hu","submitted_at":"2024-09-03T17:25:54Z","abstract_excerpt":"Current benchmarks like Needle-in-a-Haystack (NIAH), Ruler, and Needlebench focus on models' ability to understand long-context input sequences but fail to capture a critical dimension: the generation of high-quality long-form text. Applications such as design proposals, technical documentation, and creative writing rely on coherent, instruction-following outputs over extended sequences - a challenge that existing benchmarks do not adequately address. To fill this gap, we introduce LongGenBench, a novel benchmark designed to rigorously evaluate large language models' (LLMs) ability to generate"},"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":"2409.02076","kind":"arxiv","version":7},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-03T17:25:54Z","cross_cats_sorted":[],"title_canon_sha256":"81934ffb77eb51c10e9e7a5e0005ddd52fdcd317106ae6d64277bce3579a886f","abstract_canon_sha256":"5571a553f3bbf447c18b7b6984f410be5b6babf8bec804c34a07cf1d49ed9239"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:04:17.348239Z","signature_b64":"6yr9TmyyY+ddpXxCeypKU1LvQGH4NAhhx/SrZzBJEnyjFMIcPFfQs5wittGcV+23yMEWFYAg5ke+pXUoxElkCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bc9801d041f28701a574e9f3bce591bffa1a9802383c425e10c3040738f0651a","last_reissued_at":"2026-07-05T10:04:17.347788Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:04:17.347788Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LongGenBench: Benchmarking Long-Form Generation in Long Context LLMs","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ming Shan Hee, Roy Ka-Wei Lee, Yuhao Wu, Zhiqing Hu","submitted_at":"2024-09-03T17:25:54Z","abstract_excerpt":"Current benchmarks like Needle-in-a-Haystack (NIAH), Ruler, and Needlebench focus on models' ability to understand long-context input sequences but fail to capture a critical dimension: the generation of high-quality long-form text. Applications such as design proposals, technical documentation, and creative writing rely on coherent, instruction-following outputs over extended sequences - a challenge that existing benchmarks do not adequately address. To fill this gap, we introduce LongGenBench, a novel benchmark designed to rigorously evaluate large language models' (LLMs) ability to generate"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.02076","kind":"arxiv","version":7},"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/2409.02076/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":"2409.02076","created_at":"2026-07-05T10:04:17.347842+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.02076v7","created_at":"2026-07-05T10:04:17.347842+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.02076","created_at":"2026-07-05T10:04:17.347842+00:00"},{"alias_kind":"pith_short_12","alias_value":"XSMADUCB6KDQ","created_at":"2026-07-05T10:04:17.347842+00:00"},{"alias_kind":"pith_short_16","alias_value":"XSMADUCB6KDQDJLU","created_at":"2026-07-05T10:04:17.347842+00:00"},{"alias_kind":"pith_short_8","alias_value":"XSMADUCB","created_at":"2026-07-05T10:04:17.347842+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.06256","citing_title":"RedKnot: Efficient Long-Context LLM Serving with Head-Aware KV Reuse and SegPagedAttention","ref_index":66,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02028","citing_title":"Language models fail at extended rule following","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17613","citing_title":"VeriCache: Turning Lossy KV Cache into Lossless LLM Inference","ref_index":67,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10539","citing_title":"IceCache: Memory-efficient KV-cache Management for Long-Sequence LLMs","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02028","citing_title":"Language models fail at extended rule following","ref_index":19,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XSMADUCB6KDQDJLU5HZ3ZZMRX7","json":"https://pith.science/pith/XSMADUCB6KDQDJLU5HZ3ZZMRX7.json","graph_json":"https://pith.science/api/pith-number/XSMADUCB6KDQDJLU5HZ3ZZMRX7/graph.json","events_json":"https://pith.science/api/pith-number/XSMADUCB6KDQDJLU5HZ3ZZMRX7/events.json","paper":"https://pith.science/paper/XSMADUCB"},"agent_actions":{"view_html":"https://pith.science/pith/XSMADUCB6KDQDJLU5HZ3ZZMRX7","download_json":"https://pith.science/pith/XSMADUCB6KDQDJLU5HZ3ZZMRX7.json","view_paper":"https://pith.science/paper/XSMADUCB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.02076&json=true","fetch_graph":"https://pith.science/api/pith-number/XSMADUCB6KDQDJLU5HZ3ZZMRX7/graph.json","fetch_events":"https://pith.science/api/pith-number/XSMADUCB6KDQDJLU5HZ3ZZMRX7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XSMADUCB6KDQDJLU5HZ3ZZMRX7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XSMADUCB6KDQDJLU5HZ3ZZMRX7/action/storage_attestation","attest_author":"https://pith.science/pith/XSMADUCB6KDQDJLU5HZ3ZZMRX7/action/author_attestation","sign_citation":"https://pith.science/pith/XSMADUCB6KDQDJLU5HZ3ZZMRX7/action/citation_signature","submit_replication":"https://pith.science/pith/XSMADUCB6KDQDJLU5HZ3ZZMRX7/action/replication_record"}},"created_at":"2026-07-05T10:04:17.347842+00:00","updated_at":"2026-07-05T10:04:17.347842+00:00"}