{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:JO5N76H6EQHCZV3RASJZWQCBRP","short_pith_number":"pith:JO5N76H6","schema_version":"1.0","canonical_sha256":"4bbadff8fe240e2cd77104939b40418be0dc5c924ab7264e0ad55ab70e71f5df","source":{"kind":"arxiv","id":"2605.30000","version":1},"attestation_state":"computed","paper":{"title":"Cookie-Bench: Continuous On-screen Key Interaction Evaluation for Web Generation","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Fan Ding, Hangting Lou, Haoqing Yu, Haoyue Yang, Hua Wu, Jing Liu, Jingyao Li, Siqi Bao, Yifeng Kou, Zhangxiao Shen, Zhengfan Wu","submitted_at":"2026-05-28T14:30:33Z","abstract_excerpt":"Front-end web code has become a core product surface for every frontier LLM release, yet evaluating these interactive applications at development speed remains costly because human-judged leaderboards like Arena do not scale. Existing automated proxies typically lean on reference implementations, test suites, or rigid checklists, and tend to miss the reasoned synthesis a human reviewer performs over a live session. We articulate a new evaluation regime that is simultaneously reference-free, autonomously driven, and holistically reasoned, and instantiate it through two artifacts. \\textbf{\\datan"},"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":"2605.30000","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-28T14:30:33Z","cross_cats_sorted":[],"title_canon_sha256":"544b2e30faaad0e943d5e62541602298ab844fefa670b7533677f93216bf4ffb","abstract_canon_sha256":"53b3212211a2eda3cd872070f4663f40906bc475bb3cf821769e25bd0dfff1df"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T02:06:06.250245Z","signature_b64":"iXCNXJuDbnxqSTtvUELfbUPWkXqA1sXfs4U9i8B2OVI3zbMjVL+ITCXa0+OZKRSBTWYEQpKD6OyCD0N6B1lTAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4bbadff8fe240e2cd77104939b40418be0dc5c924ab7264e0ad55ab70e71f5df","last_reissued_at":"2026-05-29T02:06:06.249311Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T02:06:06.249311Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Cookie-Bench: Continuous On-screen Key Interaction Evaluation for Web Generation","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Fan Ding, Hangting Lou, Haoqing Yu, Haoyue Yang, Hua Wu, Jing Liu, Jingyao Li, Siqi Bao, Yifeng Kou, Zhangxiao Shen, Zhengfan Wu","submitted_at":"2026-05-28T14:30:33Z","abstract_excerpt":"Front-end web code has become a core product surface for every frontier LLM release, yet evaluating these interactive applications at development speed remains costly because human-judged leaderboards like Arena do not scale. Existing automated proxies typically lean on reference implementations, test suites, or rigid checklists, and tend to miss the reasoned synthesis a human reviewer performs over a live session. We articulate a new evaluation regime that is simultaneously reference-free, autonomously driven, and holistically reasoned, and instantiate it through two artifacts. \\textbf{\\datan"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.30000","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/2605.30000/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":"2605.30000","created_at":"2026-05-29T02:06:06.249495+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.30000v1","created_at":"2026-05-29T02:06:06.249495+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.30000","created_at":"2026-05-29T02:06:06.249495+00:00"},{"alias_kind":"pith_short_12","alias_value":"JO5N76H6EQHC","created_at":"2026-05-29T02:06:06.249495+00:00"},{"alias_kind":"pith_short_16","alias_value":"JO5N76H6EQHCZV3R","created_at":"2026-05-29T02:06:06.249495+00:00"},{"alias_kind":"pith_short_8","alias_value":"JO5N76H6","created_at":"2026-05-29T02:06:06.249495+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JO5N76H6EQHCZV3RASJZWQCBRP","json":"https://pith.science/pith/JO5N76H6EQHCZV3RASJZWQCBRP.json","graph_json":"https://pith.science/api/pith-number/JO5N76H6EQHCZV3RASJZWQCBRP/graph.json","events_json":"https://pith.science/api/pith-number/JO5N76H6EQHCZV3RASJZWQCBRP/events.json","paper":"https://pith.science/paper/JO5N76H6"},"agent_actions":{"view_html":"https://pith.science/pith/JO5N76H6EQHCZV3RASJZWQCBRP","download_json":"https://pith.science/pith/JO5N76H6EQHCZV3RASJZWQCBRP.json","view_paper":"https://pith.science/paper/JO5N76H6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.30000&json=true","fetch_graph":"https://pith.science/api/pith-number/JO5N76H6EQHCZV3RASJZWQCBRP/graph.json","fetch_events":"https://pith.science/api/pith-number/JO5N76H6EQHCZV3RASJZWQCBRP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JO5N76H6EQHCZV3RASJZWQCBRP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JO5N76H6EQHCZV3RASJZWQCBRP/action/storage_attestation","attest_author":"https://pith.science/pith/JO5N76H6EQHCZV3RASJZWQCBRP/action/author_attestation","sign_citation":"https://pith.science/pith/JO5N76H6EQHCZV3RASJZWQCBRP/action/citation_signature","submit_replication":"https://pith.science/pith/JO5N76H6EQHCZV3RASJZWQCBRP/action/replication_record"}},"created_at":"2026-05-29T02:06:06.249495+00:00","updated_at":"2026-05-29T02:06:06.249495+00:00"}