{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:RSD5FDDU5HITB7RDVZUEKBECT2","short_pith_number":"pith:RSD5FDDU","canonical_record":{"source":{"id":"2406.12373","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-18T07:58:33Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"a08c8d164bed182dbe42b68826b418c245abac6234ec1bdd4b3a065a2fa214b9","abstract_canon_sha256":"d9743c3cc6b9480d307ff37f77945dee3724a0cb4f27e278e62ff7460d7f5719"},"schema_version":"1.0"},"canonical_sha256":"8c87d28c74e9d130fe23ae684504829e99c881c8f6b46e564fd69c0778e2e2a7","source":{"kind":"arxiv","id":"2406.12373","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.12373","created_at":"2026-05-20T11:23:51Z"},{"alias_kind":"arxiv_version","alias_value":"2406.12373v3","created_at":"2026-05-20T11:23:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.12373","created_at":"2026-05-20T11:23:51Z"},{"alias_kind":"pith_short_12","alias_value":"RSD5FDDU5HIT","created_at":"2026-05-20T11:23:51Z"},{"alias_kind":"pith_short_16","alias_value":"RSD5FDDU5HITB7RD","created_at":"2026-05-20T11:23:51Z"},{"alias_kind":"pith_short_8","alias_value":"RSD5FDDU","created_at":"2026-05-20T11:23:51Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:RSD5FDDU5HITB7RDVZUEKBECT2","target":"record","payload":{"canonical_record":{"source":{"id":"2406.12373","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-18T07:58:33Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"a08c8d164bed182dbe42b68826b418c245abac6234ec1bdd4b3a065a2fa214b9","abstract_canon_sha256":"d9743c3cc6b9480d307ff37f77945dee3724a0cb4f27e278e62ff7460d7f5719"},"schema_version":"1.0"},"canonical_sha256":"8c87d28c74e9d130fe23ae684504829e99c881c8f6b46e564fd69c0778e2e2a7","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T11:23:51.087480Z","signature_b64":"dXYf/hFbRoAbQGYMacqduVRRPxamlZkX/xAWXlFcldVKzGepqsu8htgiyCBCHo4YTI+Ex54JxrW2tUtxSUH1CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8c87d28c74e9d130fe23ae684504829e99c881c8f6b46e564fd69c0778e2e2a7","last_reissued_at":"2026-05-20T11:23:51.085608Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T11:23:51.085608Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.12373","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T11:23:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"czPUS4ObosuZyNZjpfhsIFqcfN+6trJIzcnUgyyhZ0iyPHAkxMatJUPssWop00Cp500yhvEGXvLKVmIol5KCAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-21T04:09:59.262745Z"},"content_sha256":"505b44492f88bcf996063c20d840d46f608d459225866233958e6739d0256e73","schema_version":"1.0","event_id":"sha256:505b44492f88bcf996063c20d840d46f608d459225866233958e6739d0256e73"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:RSD5FDDU5HITB7RDVZUEKBECT2","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"WebCanvas: Benchmarking Web Agents in Online Environments","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Bing Jiang, Cheng Cui, Dehan Kong, Hangyu Liu, Shuyan Zhou, Sida Zhou, Tongshuang Wu, Yanyi Shang, Yichen Pan, Yifei Leng, Zhengyang Wu","submitted_at":"2024-06-18T07:58:33Z","abstract_excerpt":"For web agents to be practically useful, they must adapt to the continuously evolving web environment characterized by frequent updates to user interfaces and content. However, most existing benchmarks only capture the static aspects of the web. To bridge this gap, we introduce WebCanvas, an innovative online evaluation framework for web agents that effectively addresses the dynamic nature of web interactions. WebCanvas contains three main components to facilitate realistic assessments: (1) A novel evaluation metric which reliably capture critical intermediate actions or states necessary for t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.12373","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/2406.12373/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":"2026-05-20T11:23:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qTrqxg4X26oK2nvEKa3sGTrPN0/i6egIu8L3Cx8KkidfMXpuNgAOB2MmmuhQIUGeE7bdDLKLiLb5L58Juj6hDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-21T04:09:59.263120Z"},"content_sha256":"8399bb485e2463db639e749e71a6408922203cc8f7075b5e9299be420416120b","schema_version":"1.0","event_id":"sha256:8399bb485e2463db639e749e71a6408922203cc8f7075b5e9299be420416120b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/RSD5FDDU5HITB7RDVZUEKBECT2/bundle.json","state_url":"https://pith.science/pith/RSD5FDDU5HITB7RDVZUEKBECT2/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/RSD5FDDU5HITB7RDVZUEKBECT2/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":"2026-05-21T04:09:59Z","links":{"resolver":"https://pith.science/pith/RSD5FDDU5HITB7RDVZUEKBECT2","bundle":"https://pith.science/pith/RSD5FDDU5HITB7RDVZUEKBECT2/bundle.json","state":"https://pith.science/pith/RSD5FDDU5HITB7RDVZUEKBECT2/state.json","well_known_bundle":"https://pith.science/.well-known/pith/RSD5FDDU5HITB7RDVZUEKBECT2/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:RSD5FDDU5HITB7RDVZUEKBECT2","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":"d9743c3cc6b9480d307ff37f77945dee3724a0cb4f27e278e62ff7460d7f5719","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-18T07:58:33Z","title_canon_sha256":"a08c8d164bed182dbe42b68826b418c245abac6234ec1bdd4b3a065a2fa214b9"},"schema_version":"1.0","source":{"id":"2406.12373","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.12373","created_at":"2026-05-20T11:23:51Z"},{"alias_kind":"arxiv_version","alias_value":"2406.12373v3","created_at":"2026-05-20T11:23:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.12373","created_at":"2026-05-20T11:23:51Z"},{"alias_kind":"pith_short_12","alias_value":"RSD5FDDU5HIT","created_at":"2026-05-20T11:23:51Z"},{"alias_kind":"pith_short_16","alias_value":"RSD5FDDU5HITB7RD","created_at":"2026-05-20T11:23:51Z"},{"alias_kind":"pith_short_8","alias_value":"RSD5FDDU","created_at":"2026-05-20T11:23:51Z"}],"graph_snapshots":[{"event_id":"sha256:8399bb485e2463db639e749e71a6408922203cc8f7075b5e9299be420416120b","target":"graph","created_at":"2026-05-20T11:23:51Z","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/2406.12373/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"For web agents to be practically useful, they must adapt to the continuously evolving web environment characterized by frequent updates to user interfaces and content. However, most existing benchmarks only capture the static aspects of the web. To bridge this gap, we introduce WebCanvas, an innovative online evaluation framework for web agents that effectively addresses the dynamic nature of web interactions. WebCanvas contains three main components to facilitate realistic assessments: (1) A novel evaluation metric which reliably capture critical intermediate actions or states necessary for t","authors_text":"Bing Jiang, Cheng Cui, Dehan Kong, Hangyu Liu, Shuyan Zhou, Sida Zhou, Tongshuang Wu, Yanyi Shang, Yichen Pan, Yifei Leng, Zhengyang Wu","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-18T07:58:33Z","title":"WebCanvas: Benchmarking Web Agents in Online Environments"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.12373","kind":"arxiv","version":3},"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":"sha256:505b44492f88bcf996063c20d840d46f608d459225866233958e6739d0256e73","target":"record","created_at":"2026-05-20T11:23:51Z","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":"d9743c3cc6b9480d307ff37f77945dee3724a0cb4f27e278e62ff7460d7f5719","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-18T07:58:33Z","title_canon_sha256":"a08c8d164bed182dbe42b68826b418c245abac6234ec1bdd4b3a065a2fa214b9"},"schema_version":"1.0","source":{"id":"2406.12373","kind":"arxiv","version":3}},"canonical_sha256":"8c87d28c74e9d130fe23ae684504829e99c881c8f6b46e564fd69c0778e2e2a7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8c87d28c74e9d130fe23ae684504829e99c881c8f6b46e564fd69c0778e2e2a7","first_computed_at":"2026-05-20T11:23:51.085608Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T11:23:51.085608Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"dXYf/hFbRoAbQGYMacqduVRRPxamlZkX/xAWXlFcldVKzGepqsu8htgiyCBCHo4YTI+Ex54JxrW2tUtxSUH1CA==","signature_status":"signed_v1","signed_at":"2026-05-20T11:23:51.087480Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.12373","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:505b44492f88bcf996063c20d840d46f608d459225866233958e6739d0256e73","sha256:8399bb485e2463db639e749e71a6408922203cc8f7075b5e9299be420416120b"],"state_sha256":"9e7619debee537c8448e4a4cd5816166b8b8b632688385ea19e1b95e64851024"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1echXrmIwbMgBUu1yQa72LuFHoDxZ+fzIuQ1KtY0v4fJLwHzrfAj4lAFTITBV18l7op6adcLDuAv+OmEGkpbAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-21T04:09:59.265198Z","bundle_sha256":"af71d9eeaf2b685aa4d8467dae63c93d2ea04a5233eddaaadcb2def03259d052"}}