{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:JO5N76H6EQHCZV3RASJZWQCBRP","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"53b3212211a2eda3cd872070f4663f40906bc475bb3cf821769e25bd0dfff1df","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-28T14:30:33Z","title_canon_sha256":"544b2e30faaad0e943d5e62541602298ab844fefa670b7533677f93216bf4ffb"},"schema_version":"1.0","source":{"id":"2605.30000","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.30000","created_at":"2026-05-29T02:06:06Z"},{"alias_kind":"arxiv_version","alias_value":"2605.30000v1","created_at":"2026-05-29T02:06:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.30000","created_at":"2026-05-29T02:06:06Z"},{"alias_kind":"pith_short_12","alias_value":"JO5N76H6EQHC","created_at":"2026-05-29T02:06:06Z"},{"alias_kind":"pith_short_16","alias_value":"JO5N76H6EQHCZV3R","created_at":"2026-05-29T02:06:06Z"},{"alias_kind":"pith_short_8","alias_value":"JO5N76H6","created_at":"2026-05-29T02:06:06Z"}],"graph_snapshots":[{"event_id":"sha256:03da5d4e785ef68a25e6b5150386f5bb12e4ca5598bd8b4de5f4288dd1eac8b0","target":"graph","created_at":"2026-05-29T02:06:06Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2605.30000/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Fan Ding, Hangting Lou, Haoqing Yu, Haoyue Yang, Hua Wu, Jing Liu, Jingyao Li, Siqi Bao, Yifeng Kou, Zhangxiao Shen, Zhengfan Wu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-28T14:30:33Z","title":"Cookie-Bench: Continuous On-screen Key Interaction Evaluation for Web Generation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.30000","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:31838a9d7f3a3dcf80eee03bf96cc16fa7c10d27836696a41c883c784db4c417","target":"record","created_at":"2026-05-29T02:06:06Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"53b3212211a2eda3cd872070f4663f40906bc475bb3cf821769e25bd0dfff1df","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-28T14:30:33Z","title_canon_sha256":"544b2e30faaad0e943d5e62541602298ab844fefa670b7533677f93216bf4ffb"},"schema_version":"1.0","source":{"id":"2605.30000","kind":"arxiv","version":1}},"canonical_sha256":"4bbadff8fe240e2cd77104939b40418be0dc5c924ab7264e0ad55ab70e71f5df","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4bbadff8fe240e2cd77104939b40418be0dc5c924ab7264e0ad55ab70e71f5df","first_computed_at":"2026-05-29T02:06:06.249311Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-29T02:06:06.249311Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"iXCNXJuDbnxqSTtvUELfbUPWkXqA1sXfs4U9i8B2OVI3zbMjVL+ITCXa0+OZKRSBTWYEQpKD6OyCD0N6B1lTAg==","signature_status":"signed_v1","signed_at":"2026-05-29T02:06:06.250245Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.30000","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:31838a9d7f3a3dcf80eee03bf96cc16fa7c10d27836696a41c883c784db4c417","sha256:03da5d4e785ef68a25e6b5150386f5bb12e4ca5598bd8b4de5f4288dd1eac8b0"],"state_sha256":"6e5224f25aa1255fb358fafdbc2b1606b1ec8a5d837d73c75c444e146c36588b"}