{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:KIZ6EHJI4CJKYIILACXFFUDY5T","short_pith_number":"pith:KIZ6EHJI","canonical_record":{"source":{"id":"2608.03689","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-04T13:57:48Z","cross_cats_sorted":["cs.SE"],"title_canon_sha256":"3de76091a6e409e789da6477301e3acd6a88d5af2554c2f4bbb2086716164321","abstract_canon_sha256":"9d6397b40434cbac63fd29dc1f08f751cf0b4992430b812fcb171cde73091083"},"schema_version":"1.0"},"canonical_sha256":"5233e21d28e092ac210b00ae52d078ecc6bcbe7b4cf1ccb612c9a88f4560f7a2","source":{"kind":"arxiv","id":"2608.03689","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.03689","created_at":"2026-08-05T01:36:54Z"},{"alias_kind":"arxiv_version","alias_value":"2608.03689v1","created_at":"2026-08-05T01:36:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.03689","created_at":"2026-08-05T01:36:54Z"},{"alias_kind":"pith_short_12","alias_value":"KIZ6EHJI4CJK","created_at":"2026-08-05T01:36:54Z"},{"alias_kind":"pith_short_16","alias_value":"KIZ6EHJI4CJKYIIL","created_at":"2026-08-05T01:36:54Z"},{"alias_kind":"pith_short_8","alias_value":"KIZ6EHJI","created_at":"2026-08-05T01:36:54Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:KIZ6EHJI4CJKYIILACXFFUDY5T","target":"record","payload":{"canonical_record":{"source":{"id":"2608.03689","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-04T13:57:48Z","cross_cats_sorted":["cs.SE"],"title_canon_sha256":"3de76091a6e409e789da6477301e3acd6a88d5af2554c2f4bbb2086716164321","abstract_canon_sha256":"9d6397b40434cbac63fd29dc1f08f751cf0b4992430b812fcb171cde73091083"},"schema_version":"1.0"},"canonical_sha256":"5233e21d28e092ac210b00ae52d078ecc6bcbe7b4cf1ccb612c9a88f4560f7a2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-05T01:36:54.016370Z","signature_b64":"piq5KrAAiJtNOOwDNiuClaaSF2HjHL5s1mpNL4709WjBbIvmvWK0B5gN39q/Ba1lNcylDzxarm71SbrO+BktAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5233e21d28e092ac210b00ae52d078ecc6bcbe7b4cf1ccb612c9a88f4560f7a2","last_reissued_at":"2026-08-05T01:36:54.014967Z","signature_status":"signed_v1","first_computed_at":"2026-08-05T01:36:54.014967Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2608.03689","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-05T01:36:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0+dEY0xzTtcjaop5p33xJOk1PQT30skCCuY5yy7iTSZvMgZTdI9XC8uYzSh2WB73CxcJ1GGR0wnEPmL1TAm1BA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T10:53:41.639350Z"},"content_sha256":"d0301b1682881ca94497811787264172be4fc38fdacd049c0582c031a661c85f","schema_version":"1.0","event_id":"sha256:d0301b1682881ca94497811787264172be4fc38fdacd049c0582c031a661c85f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:KIZ6EHJI4CJKYIILACXFFUDY5T","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"LiveEvalBench: Toward Open-World Evaluation for Web Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SE"],"primary_cat":"cs.AI","authors_text":"Jun Zhou, Lin Yuan, Wei Chen, Xiaolau Zhang, Yinghao Tang, Yixiao Fu, Yiyao Wang, Zhen Wen","submitted_at":"2026-08-04T13:57:48Z","abstract_excerpt":"Large language models are increasingly capable of synthesizing executable frontend projects, yet existing benchmarks still treat web generation as a static evaluation problem. We argue that frontend artifacts demand a different paradigm: they are interactive rather than static, admit diverse yet equally valid implementations, and evolve faster than rigid pipelines can accommodate. To address these gaps, we present LiveEvalBench, an automated framework that reformulates web-generation evaluation as an agentic, adaptive, and extensible process. LiveEvalBench instantiates evaluation as a collabor"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.03689","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/2608.03689/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-05T01:36:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RXr6Ipn3bI6jA5wPBblchjTkrSZvHQX9b7ZTOA8hsYiO2wOiBzQEDgg9f/hE+dZKUFuBOwVg1J2Vt1KtRalHBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T10:53:41.639960Z"},"content_sha256":"b71434821d4763f4beee90be5d2f2874f8565a0dc299741974f64a0f1bf74301","schema_version":"1.0","event_id":"sha256:b71434821d4763f4beee90be5d2f2874f8565a0dc299741974f64a0f1bf74301"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KIZ6EHJI4CJKYIILACXFFUDY5T/bundle.json","state_url":"https://pith.science/pith/KIZ6EHJI4CJKYIILACXFFUDY5T/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KIZ6EHJI4CJKYIILACXFFUDY5T/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-10T10:53:41Z","links":{"resolver":"https://pith.science/pith/KIZ6EHJI4CJKYIILACXFFUDY5T","bundle":"https://pith.science/pith/KIZ6EHJI4CJKYIILACXFFUDY5T/bundle.json","state":"https://pith.science/pith/KIZ6EHJI4CJKYIILACXFFUDY5T/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KIZ6EHJI4CJKYIILACXFFUDY5T/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:KIZ6EHJI4CJKYIILACXFFUDY5T","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":"9d6397b40434cbac63fd29dc1f08f751cf0b4992430b812fcb171cde73091083","cross_cats_sorted":["cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-04T13:57:48Z","title_canon_sha256":"3de76091a6e409e789da6477301e3acd6a88d5af2554c2f4bbb2086716164321"},"schema_version":"1.0","source":{"id":"2608.03689","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.03689","created_at":"2026-08-05T01:36:54Z"},{"alias_kind":"arxiv_version","alias_value":"2608.03689v1","created_at":"2026-08-05T01:36:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.03689","created_at":"2026-08-05T01:36:54Z"},{"alias_kind":"pith_short_12","alias_value":"KIZ6EHJI4CJK","created_at":"2026-08-05T01:36:54Z"},{"alias_kind":"pith_short_16","alias_value":"KIZ6EHJI4CJKYIIL","created_at":"2026-08-05T01:36:54Z"},{"alias_kind":"pith_short_8","alias_value":"KIZ6EHJI","created_at":"2026-08-05T01:36:54Z"}],"graph_snapshots":[{"event_id":"sha256:b71434821d4763f4beee90be5d2f2874f8565a0dc299741974f64a0f1bf74301","target":"graph","created_at":"2026-08-05T01:36:54Z","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/2608.03689/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models are increasingly capable of synthesizing executable frontend projects, yet existing benchmarks still treat web generation as a static evaluation problem. We argue that frontend artifacts demand a different paradigm: they are interactive rather than static, admit diverse yet equally valid implementations, and evolve faster than rigid pipelines can accommodate. To address these gaps, we present LiveEvalBench, an automated framework that reformulates web-generation evaluation as an agentic, adaptive, and extensible process. LiveEvalBench instantiates evaluation as a collabor","authors_text":"Jun Zhou, Lin Yuan, Wei Chen, Xiaolau Zhang, Yinghao Tang, Yixiao Fu, Yiyao Wang, Zhen Wen","cross_cats":["cs.SE"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-04T13:57:48Z","title":"LiveEvalBench: Toward Open-World Evaluation for Web Generation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.03689","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:d0301b1682881ca94497811787264172be4fc38fdacd049c0582c031a661c85f","target":"record","created_at":"2026-08-05T01:36:54Z","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":"9d6397b40434cbac63fd29dc1f08f751cf0b4992430b812fcb171cde73091083","cross_cats_sorted":["cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-04T13:57:48Z","title_canon_sha256":"3de76091a6e409e789da6477301e3acd6a88d5af2554c2f4bbb2086716164321"},"schema_version":"1.0","source":{"id":"2608.03689","kind":"arxiv","version":1}},"canonical_sha256":"5233e21d28e092ac210b00ae52d078ecc6bcbe7b4cf1ccb612c9a88f4560f7a2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5233e21d28e092ac210b00ae52d078ecc6bcbe7b4cf1ccb612c9a88f4560f7a2","first_computed_at":"2026-08-05T01:36:54.014967Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-05T01:36:54.014967Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"piq5KrAAiJtNOOwDNiuClaaSF2HjHL5s1mpNL4709WjBbIvmvWK0B5gN39q/Ba1lNcylDzxarm71SbrO+BktAA==","signature_status":"signed_v1","signed_at":"2026-08-05T01:36:54.016370Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.03689","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d0301b1682881ca94497811787264172be4fc38fdacd049c0582c031a661c85f","sha256:b71434821d4763f4beee90be5d2f2874f8565a0dc299741974f64a0f1bf74301"],"state_sha256":"8f9474d27897d27e2e1fa829ea370971986ed35f03f81beee2a89595af050df2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Lry6soTied2shvj4FFVXo2+HkSN0nSkgGek8cFhq36+FRhLJHgf1fF7wF/e2RjgMihfNzWVUiY5BiTsHhu3WCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T10:53:41.644636Z","bundle_sha256":"f85c861fd119abde68c8ffbd0253c20db9072f98207e2c9004b59ecb6ddfd9f0"}}