{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:XZEJ6E56JKISFVZ5563WRTAODH","short_pith_number":"pith:XZEJ6E56","canonical_record":{"source":{"id":"2403.11905","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-03-18T16:06:30Z","cross_cats_sorted":["cs.CL","cs.CV","cs.HC"],"title_canon_sha256":"9ca0a2b37d57a991f9444896ea8d2c609f369417baff6759a9d97205af40c59c","abstract_canon_sha256":"c0d7bdb631dcd0f93dda6ed514d9dd303865a23189b6b93c24538bc6385fabad"},"schema_version":"1.0"},"canonical_sha256":"be489f13be4a9122d73defb768cc0e19dd29e7eedcf4659042f49ddddff94f6e","source":{"kind":"arxiv","id":"2403.11905","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.11905","created_at":"2026-07-05T10:18:08Z"},{"alias_kind":"arxiv_version","alias_value":"2403.11905v4","created_at":"2026-07-05T10:18:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.11905","created_at":"2026-07-05T10:18:08Z"},{"alias_kind":"pith_short_12","alias_value":"XZEJ6E56JKIS","created_at":"2026-07-05T10:18:08Z"},{"alias_kind":"pith_short_16","alias_value":"XZEJ6E56JKISFVZ5","created_at":"2026-07-05T10:18:08Z"},{"alias_kind":"pith_short_8","alias_value":"XZEJ6E56","created_at":"2026-07-05T10:18:08Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:XZEJ6E56JKISFVZ5563WRTAODH","target":"record","payload":{"canonical_record":{"source":{"id":"2403.11905","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-03-18T16:06:30Z","cross_cats_sorted":["cs.CL","cs.CV","cs.HC"],"title_canon_sha256":"9ca0a2b37d57a991f9444896ea8d2c609f369417baff6759a9d97205af40c59c","abstract_canon_sha256":"c0d7bdb631dcd0f93dda6ed514d9dd303865a23189b6b93c24538bc6385fabad"},"schema_version":"1.0"},"canonical_sha256":"be489f13be4a9122d73defb768cc0e19dd29e7eedcf4659042f49ddddff94f6e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:18:08.003795Z","signature_b64":"Bz/uUO5cwSONb7xw0if65DNA1Ah3bsloT1O/VFn5a80PeHiALlJ2sviZPwydlmjf3+qGegy8vn1jU3dK7jb8Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"be489f13be4a9122d73defb768cc0e19dd29e7eedcf4659042f49ddddff94f6e","last_reissued_at":"2026-07-05T10:18:08.003386Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:18:08.003386Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2403.11905","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:18:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1/WnUcJZAPvjxsdiSpo/yTqMamwNFI061uXY3S0XNorhaGiIUOlWhtlwBi3Ur346Nyq8zc0yQZit78st9cTLBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T17:49:30.642215Z"},"content_sha256":"531b581e1c13f871c902e5341eaa7065b5ba549ccdf64631b29706f6f43c902d","schema_version":"1.0","event_id":"sha256:531b581e1c13f871c902e5341eaa7065b5ba549ccdf64631b29706f6f43c902d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:XZEJ6E56JKISFVZ5563WRTAODH","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Tur[k]ingBench: A Challenge Benchmark for Web Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.CV","cs.HC"],"primary_cat":"cs.AI","authors_text":"Adam Byerly, Adi Asija, Benjamin Van Durme, Daniel Khashabi, Jingyu Zhang, Kate Sanders, Kevin Xu, Tanay Nayak, Yeganeh Kordi, Yizhong Wang","submitted_at":"2024-03-18T16:06:30Z","abstract_excerpt":"Can advanced multi-modal models effectively tackle complex web-based tasks? Such tasks are often found on crowdsourcing platforms, where crowdworkers engage in challenging micro-tasks within web-based environments.\n  Building on this idea, we present TurkingBench, a benchmark consisting of tasks presented as web pages with textual instructions and multi-modal contexts. Unlike previous approaches that rely on artificially synthesized web pages, our benchmark uses natural HTML pages originally designed for crowdsourcing workers to perform various annotation tasks. Each task's HTML instructions a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.11905","kind":"arxiv","version":4},"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/2403.11905/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":"2026-07-05T10:18:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4Ofj4jpdhFy527ZJEdZvwJbyikXumZ6TxMzbeRNWPPnLzQLx0HRic/lZNVzEdvGqBBGgbthaaFY/zsDFbkjICg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T17:49:30.642708Z"},"content_sha256":"551f7688a143203b1559c375137db44ff7a8a26be2c9dd55e7950ca15d006381","schema_version":"1.0","event_id":"sha256:551f7688a143203b1559c375137db44ff7a8a26be2c9dd55e7950ca15d006381"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XZEJ6E56JKISFVZ5563WRTAODH/bundle.json","state_url":"https://pith.science/pith/XZEJ6E56JKISFVZ5563WRTAODH/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XZEJ6E56JKISFVZ5563WRTAODH/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":"2026-08-11T17:49:30Z","links":{"resolver":"https://pith.science/pith/XZEJ6E56JKISFVZ5563WRTAODH","bundle":"https://pith.science/pith/XZEJ6E56JKISFVZ5563WRTAODH/bundle.json","state":"https://pith.science/pith/XZEJ6E56JKISFVZ5563WRTAODH/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XZEJ6E56JKISFVZ5563WRTAODH/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:XZEJ6E56JKISFVZ5563WRTAODH","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":"c0d7bdb631dcd0f93dda6ed514d9dd303865a23189b6b93c24538bc6385fabad","cross_cats_sorted":["cs.CL","cs.CV","cs.HC"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-03-18T16:06:30Z","title_canon_sha256":"9ca0a2b37d57a991f9444896ea8d2c609f369417baff6759a9d97205af40c59c"},"schema_version":"1.0","source":{"id":"2403.11905","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.11905","created_at":"2026-07-05T10:18:08Z"},{"alias_kind":"arxiv_version","alias_value":"2403.11905v4","created_at":"2026-07-05T10:18:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.11905","created_at":"2026-07-05T10:18:08Z"},{"alias_kind":"pith_short_12","alias_value":"XZEJ6E56JKIS","created_at":"2026-07-05T10:18:08Z"},{"alias_kind":"pith_short_16","alias_value":"XZEJ6E56JKISFVZ5","created_at":"2026-07-05T10:18:08Z"},{"alias_kind":"pith_short_8","alias_value":"XZEJ6E56","created_at":"2026-07-05T10:18:08Z"}],"graph_snapshots":[{"event_id":"sha256:551f7688a143203b1559c375137db44ff7a8a26be2c9dd55e7950ca15d006381","target":"graph","created_at":"2026-07-05T10:18:08Z","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/2403.11905/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Can advanced multi-modal models effectively tackle complex web-based tasks? Such tasks are often found on crowdsourcing platforms, where crowdworkers engage in challenging micro-tasks within web-based environments.\n  Building on this idea, we present TurkingBench, a benchmark consisting of tasks presented as web pages with textual instructions and multi-modal contexts. Unlike previous approaches that rely on artificially synthesized web pages, our benchmark uses natural HTML pages originally designed for crowdsourcing workers to perform various annotation tasks. Each task's HTML instructions a","authors_text":"Adam Byerly, Adi Asija, Benjamin Van Durme, Daniel Khashabi, Jingyu Zhang, Kate Sanders, Kevin Xu, Tanay Nayak, Yeganeh Kordi, Yizhong Wang","cross_cats":["cs.CL","cs.CV","cs.HC"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-03-18T16:06:30Z","title":"Tur[k]ingBench: A Challenge Benchmark for Web Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.11905","kind":"arxiv","version":4},"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":"sha256:531b581e1c13f871c902e5341eaa7065b5ba549ccdf64631b29706f6f43c902d","target":"record","created_at":"2026-07-05T10:18:08Z","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":"c0d7bdb631dcd0f93dda6ed514d9dd303865a23189b6b93c24538bc6385fabad","cross_cats_sorted":["cs.CL","cs.CV","cs.HC"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-03-18T16:06:30Z","title_canon_sha256":"9ca0a2b37d57a991f9444896ea8d2c609f369417baff6759a9d97205af40c59c"},"schema_version":"1.0","source":{"id":"2403.11905","kind":"arxiv","version":4}},"canonical_sha256":"be489f13be4a9122d73defb768cc0e19dd29e7eedcf4659042f49ddddff94f6e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"be489f13be4a9122d73defb768cc0e19dd29e7eedcf4659042f49ddddff94f6e","first_computed_at":"2026-07-05T10:18:08.003386Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:18:08.003386Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Bz/uUO5cwSONb7xw0if65DNA1Ah3bsloT1O/VFn5a80PeHiALlJ2sviZPwydlmjf3+qGegy8vn1jU3dK7jb8Dw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:18:08.003795Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.11905","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:531b581e1c13f871c902e5341eaa7065b5ba549ccdf64631b29706f6f43c902d","sha256:551f7688a143203b1559c375137db44ff7a8a26be2c9dd55e7950ca15d006381"],"state_sha256":"36c67572dd3f1171d44d8ebace4f987f3651ce14a7fa4ff075824d4cd121994f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"G63/kz/Gta7HvXAvRUu6AJlgVV/QbU0qiNZy+eYnIYiebuRhfsmBjZj7g7rX6qMGrvCCqQabzMYzggFevR2ABA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T17:49:30.645947Z","bundle_sha256":"08f48ae450802bff970ea4b8dbd0f3dd34c83f9bb0d388dfa9a00536f1dc5504"}}