{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:RCRWEHATIN36YQMTW7GLV5TJFQ","short_pith_number":"pith:RCRWEHAT","schema_version":"1.0","canonical_sha256":"88a3621c134377ec4193b7ccbaf6692c2864abe974b3f30339f3ed1bde7d9c90","source":{"kind":"arxiv","id":"2504.07532","version":3},"attestation_state":"computed","paper":{"title":"AI-Slop to AI-Polish? Aligning Language Models through Edit-Based Writing Rewards and Test-time Computation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Chien-Sheng Wu, Philippe Laban, Tuhin Chakrabarty","submitted_at":"2025-04-10T07:58:05Z","abstract_excerpt":"AI-generated text is proliferating across domains, from creative writing and journalism to marketing content and scientific articles. Models can follow user-provided instructions to generate coherent and grammatically correct outputs but in this work, we study a more fundamental question: how do we evaluate and improve the writing quality of AI-generated text? Writing quality assessment has received less attention from the community, in part because it is fundamentally subjective and requires expertise. We first introduce the Writing Quality Benchmark (WQ) by consolidating five writing-prefere"},"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":"2504.07532","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-10T07:58:05Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"8b102a07cc9de30d02a168fa6912190223dc343b4498ba0de80fe5d1c8007471","abstract_canon_sha256":"18f7fa6314ed843eb88bb4df25551b10b208fa262a33ff892669f6900886a3a6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:53:01.089202Z","signature_b64":"X1jVpoC1q3C9s+mqOUw5jYdoRAGPDN8OTkZm7xRYORwLf+0ySjuaVJP+cX3STHj02p2VeW0soioiYaXrXx6qBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"88a3621c134377ec4193b7ccbaf6692c2864abe974b3f30339f3ed1bde7d9c90","last_reissued_at":"2026-07-05T11:53:01.088798Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:53:01.088798Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AI-Slop to AI-Polish? Aligning Language Models through Edit-Based Writing Rewards and Test-time Computation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Chien-Sheng Wu, Philippe Laban, Tuhin Chakrabarty","submitted_at":"2025-04-10T07:58:05Z","abstract_excerpt":"AI-generated text is proliferating across domains, from creative writing and journalism to marketing content and scientific articles. Models can follow user-provided instructions to generate coherent and grammatically correct outputs but in this work, we study a more fundamental question: how do we evaluate and improve the writing quality of AI-generated text? Writing quality assessment has received less attention from the community, in part because it is fundamentally subjective and requires expertise. We first introduce the Writing Quality Benchmark (WQ) by consolidating five writing-prefere"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.07532","kind":"arxiv","version":3},"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/2504.07532/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":"2504.07532","created_at":"2026-07-05T11:53:01.088857+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.07532v3","created_at":"2026-07-05T11:53:01.088857+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.07532","created_at":"2026-07-05T11:53:01.088857+00:00"},{"alias_kind":"pith_short_12","alias_value":"RCRWEHATIN36","created_at":"2026-07-05T11:53:01.088857+00:00"},{"alias_kind":"pith_short_16","alias_value":"RCRWEHATIN36YQMT","created_at":"2026-07-05T11:53:01.088857+00:00"},{"alias_kind":"pith_short_8","alias_value":"RCRWEHAT","created_at":"2026-07-05T11:53:01.088857+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25152","citing_title":"Hitting a Moving Target: Test-Time Adaptation for AI Text Detection under Continual Distribution Shift","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2505.06120","citing_title":"LLMs Get Lost In Multi-Turn Conversation","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06552","citing_title":"To Lie or Not to Lie? Investigating The Biased Spread of Global Lies by LLMs","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2604.07643","citing_title":"Narrix: Remixing Narrative Strategies from Examples for Story Writing","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15597","citing_title":"LLMs Corrupt Your Documents When You Delegate","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RCRWEHATIN36YQMTW7GLV5TJFQ","json":"https://pith.science/pith/RCRWEHATIN36YQMTW7GLV5TJFQ.json","graph_json":"https://pith.science/api/pith-number/RCRWEHATIN36YQMTW7GLV5TJFQ/graph.json","events_json":"https://pith.science/api/pith-number/RCRWEHATIN36YQMTW7GLV5TJFQ/events.json","paper":"https://pith.science/paper/RCRWEHAT"},"agent_actions":{"view_html":"https://pith.science/pith/RCRWEHATIN36YQMTW7GLV5TJFQ","download_json":"https://pith.science/pith/RCRWEHATIN36YQMTW7GLV5TJFQ.json","view_paper":"https://pith.science/paper/RCRWEHAT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.07532&json=true","fetch_graph":"https://pith.science/api/pith-number/RCRWEHATIN36YQMTW7GLV5TJFQ/graph.json","fetch_events":"https://pith.science/api/pith-number/RCRWEHATIN36YQMTW7GLV5TJFQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RCRWEHATIN36YQMTW7GLV5TJFQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RCRWEHATIN36YQMTW7GLV5TJFQ/action/storage_attestation","attest_author":"https://pith.science/pith/RCRWEHATIN36YQMTW7GLV5TJFQ/action/author_attestation","sign_citation":"https://pith.science/pith/RCRWEHATIN36YQMTW7GLV5TJFQ/action/citation_signature","submit_replication":"https://pith.science/pith/RCRWEHATIN36YQMTW7GLV5TJFQ/action/replication_record"}},"created_at":"2026-07-05T11:53:01.088857+00:00","updated_at":"2026-07-05T11:53:01.088857+00:00"}