{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:3IL7K6HDSUBXBNCBPO77HHXZLK","short_pith_number":"pith:3IL7K6HD","schema_version":"1.0","canonical_sha256":"da17f578e3950370b4417bbff39ef95aa0bb75f545d405a7fd296fcc97e45254","source":{"kind":"arxiv","id":"2508.18739","version":1},"attestation_state":"computed","paper":{"title":"Beyond Quality: Unlocking Diversity in Ad Headline Generation with Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Chang Wang, Depeng Yuan, Kedi Chen, Mingrui Zhu, Ruiwen Xu, Shuhao Li, Siyu Yan, Yanhua Huang, Yinqi Zhang, Yuanhang Zheng, Yuqi Chen","submitted_at":"2025-08-26T07:11:44Z","abstract_excerpt":"The generation of ad headlines plays a vital role in modern advertising, where both quality and diversity are essential to engage a broad range of audience segments. Current approaches primarily optimize language models for headline quality or click-through rates (CTR), often overlooking the need for diversity and resulting in homogeneous outputs. To address this limitation, we propose DIVER, a novel framework based on large language models (LLMs) that are jointly optimized for both diversity and quality. We first design a semantic- and stylistic-aware data generation pipeline that automatical"},"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":"2508.18739","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-26T07:11:44Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"d8dbbf53d0b6ffc5f4856184e20f98bfe02633f2327d6136ce8f7f84cbf8a6ce","abstract_canon_sha256":"fb686962ceb7670616295cfd9f8247276f05f9ad34801a2d8afad60071d7c0c4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:59:30.001834Z","signature_b64":"QNeSJS8MRApTFVcV21P/qrEmvl6OgD1L+8cwfQCktjFoG9qSNZ4sM50SRfmq5+BGXKU+c/EAT6LtZOJzjOViDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"da17f578e3950370b4417bbff39ef95aa0bb75f545d405a7fd296fcc97e45254","last_reissued_at":"2026-07-05T11:59:30.001370Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:59:30.001370Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Quality: Unlocking Diversity in Ad Headline Generation with Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Chang Wang, Depeng Yuan, Kedi Chen, Mingrui Zhu, Ruiwen Xu, Shuhao Li, Siyu Yan, Yanhua Huang, Yinqi Zhang, Yuanhang Zheng, Yuqi Chen","submitted_at":"2025-08-26T07:11:44Z","abstract_excerpt":"The generation of ad headlines plays a vital role in modern advertising, where both quality and diversity are essential to engage a broad range of audience segments. Current approaches primarily optimize language models for headline quality or click-through rates (CTR), often overlooking the need for diversity and resulting in homogeneous outputs. To address this limitation, we propose DIVER, a novel framework based on large language models (LLMs) that are jointly optimized for both diversity and quality. We first design a semantic- and stylistic-aware data generation pipeline that automatical"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.18739","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/2508.18739/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":"2508.18739","created_at":"2026-07-05T11:59:30.001430+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.18739v1","created_at":"2026-07-05T11:59:30.001430+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.18739","created_at":"2026-07-05T11:59:30.001430+00:00"},{"alias_kind":"pith_short_12","alias_value":"3IL7K6HDSUBX","created_at":"2026-07-05T11:59:30.001430+00:00"},{"alias_kind":"pith_short_16","alias_value":"3IL7K6HDSUBXBNCB","created_at":"2026-07-05T11:59:30.001430+00:00"},{"alias_kind":"pith_short_8","alias_value":"3IL7K6HD","created_at":"2026-07-05T11:59:30.001430+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.12138","citing_title":"Design Your Ad: Personalized Advertising Image and Text Generation with Unified Autoregressive Models","ref_index":61,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3IL7K6HDSUBXBNCBPO77HHXZLK","json":"https://pith.science/pith/3IL7K6HDSUBXBNCBPO77HHXZLK.json","graph_json":"https://pith.science/api/pith-number/3IL7K6HDSUBXBNCBPO77HHXZLK/graph.json","events_json":"https://pith.science/api/pith-number/3IL7K6HDSUBXBNCBPO77HHXZLK/events.json","paper":"https://pith.science/paper/3IL7K6HD"},"agent_actions":{"view_html":"https://pith.science/pith/3IL7K6HDSUBXBNCBPO77HHXZLK","download_json":"https://pith.science/pith/3IL7K6HDSUBXBNCBPO77HHXZLK.json","view_paper":"https://pith.science/paper/3IL7K6HD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.18739&json=true","fetch_graph":"https://pith.science/api/pith-number/3IL7K6HDSUBXBNCBPO77HHXZLK/graph.json","fetch_events":"https://pith.science/api/pith-number/3IL7K6HDSUBXBNCBPO77HHXZLK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3IL7K6HDSUBXBNCBPO77HHXZLK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3IL7K6HDSUBXBNCBPO77HHXZLK/action/storage_attestation","attest_author":"https://pith.science/pith/3IL7K6HDSUBXBNCBPO77HHXZLK/action/author_attestation","sign_citation":"https://pith.science/pith/3IL7K6HDSUBXBNCBPO77HHXZLK/action/citation_signature","submit_replication":"https://pith.science/pith/3IL7K6HDSUBXBNCBPO77HHXZLK/action/replication_record"}},"created_at":"2026-07-05T11:59:30.001430+00:00","updated_at":"2026-07-05T11:59:30.001430+00:00"}