{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:3ZRBUH3JDP3G4SNENE4IOHQM7B","short_pith_number":"pith:3ZRBUH3J","canonical_record":{"source":{"id":"2607.06118","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-07T10:27:31Z","cross_cats_sorted":["cs.MM"],"title_canon_sha256":"245e8a4451867a8ea4555ae2768891162eab1daa59297e4bef6bb4ba6c82e9f6","abstract_canon_sha256":"3e297988184a8344ac5fd291573c6eee3436ceaee0cd1358d4673f22baeca238"},"schema_version":"1.0"},"canonical_sha256":"de621a1f691bf66e49a46938871e0cf878446db0ebba83eb5913110b6ad1dd90","source":{"kind":"arxiv","id":"2607.06118","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.06118","created_at":"2026-07-08T01:18:57Z"},{"alias_kind":"arxiv_version","alias_value":"2607.06118v1","created_at":"2026-07-08T01:18:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.06118","created_at":"2026-07-08T01:18:57Z"},{"alias_kind":"pith_short_12","alias_value":"3ZRBUH3JDP3G","created_at":"2026-07-08T01:18:57Z"},{"alias_kind":"pith_short_16","alias_value":"3ZRBUH3JDP3G4SNE","created_at":"2026-07-08T01:18:57Z"},{"alias_kind":"pith_short_8","alias_value":"3ZRBUH3J","created_at":"2026-07-08T01:18:57Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:3ZRBUH3JDP3G4SNENE4IOHQM7B","target":"record","payload":{"canonical_record":{"source":{"id":"2607.06118","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-07T10:27:31Z","cross_cats_sorted":["cs.MM"],"title_canon_sha256":"245e8a4451867a8ea4555ae2768891162eab1daa59297e4bef6bb4ba6c82e9f6","abstract_canon_sha256":"3e297988184a8344ac5fd291573c6eee3436ceaee0cd1358d4673f22baeca238"},"schema_version":"1.0"},"canonical_sha256":"de621a1f691bf66e49a46938871e0cf878446db0ebba83eb5913110b6ad1dd90","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-08T01:18:57.414432Z","signature_b64":"cexW5WCChc5f7TvrnaIbSZKEgaoUl212fPG4BbVGYunpV/Jw4F+GOGYyItSxoG73nnmvGMo4+QzSOTNPheVSCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"de621a1f691bf66e49a46938871e0cf878446db0ebba83eb5913110b6ad1dd90","last_reissued_at":"2026-07-08T01:18:57.413939Z","signature_status":"signed_v1","first_computed_at":"2026-07-08T01:18:57.413939Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.06118","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":"2026-07-08T01:18:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GuDjEzps0kQm2LINQ+MTXFXHwPvodau2oMUFfXywWGx2NljXJeDkj7YXHnEyLIiZU++ZW29w+ll9+s8CPyEnCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-11T14:42:44.313791Z"},"content_sha256":"75e0e018f3214629298fd0d6e2049715d678bac8e50a92ef1444655182e65a2b","schema_version":"1.0","event_id":"sha256:75e0e018f3214629298fd0d6e2049715d678bac8e50a92ef1444655182e65a2b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:3ZRBUH3JDP3G4SNENE4IOHQM7B","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"WebRetriever: A Large-Scale Comprehensive Benchmark for Efficient Web Agent Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.MM"],"primary_cat":"cs.CV","authors_text":"Anyang Su, Chenxu Zhao, Hanning Wang, Minghui Wu, Ping Jiang, Shuo Wang, Tianyu Fu, Wei Dong, Yuyang Chen, Zhen Lei, Zhe Yu, Zhizhou Fang","submitted_at":"2026-07-07T10:27:31Z","abstract_excerpt":"As web agents increasingly demonstrate capabilities in automated task execution, the development of robust evaluation frameworks for assessing their navigation and task completion performance has emerged as a critical research priority. However, existing benchmarks exhibit fundamental limitations. First, they suffer from insufficient scale and limited domain diversity, constraining comprehensive evaluation of cross-domain generalization. Second, prevailing LLM-as-Judge evaluation methodologies inadequately capture fine-grained interaction semantics, particularly regarding precise query formula"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.06118","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/2607.06118/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":"2026-07-08T01:18:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XVifS7Iwuf5YKXBDmIs2NHsm84rZqZ7lUomtDkQMNqLbXCAxP3nJCTxdmmUszpaqCpjOQHMJe1oUEDPrCF6oAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-11T14:42:44.314161Z"},"content_sha256":"a96f842473793e2579001ca224afc865370fb8da44b77ae89d8b252c5d752b4c","schema_version":"1.0","event_id":"sha256:a96f842473793e2579001ca224afc865370fb8da44b77ae89d8b252c5d752b4c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/3ZRBUH3JDP3G4SNENE4IOHQM7B/bundle.json","state_url":"https://pith.science/pith/3ZRBUH3JDP3G4SNENE4IOHQM7B/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/3ZRBUH3JDP3G4SNENE4IOHQM7B/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":"2026-07-11T14:42:44Z","links":{"resolver":"https://pith.science/pith/3ZRBUH3JDP3G4SNENE4IOHQM7B","bundle":"https://pith.science/pith/3ZRBUH3JDP3G4SNENE4IOHQM7B/bundle.json","state":"https://pith.science/pith/3ZRBUH3JDP3G4SNENE4IOHQM7B/state.json","well_known_bundle":"https://pith.science/.well-known/pith/3ZRBUH3JDP3G4SNENE4IOHQM7B/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:3ZRBUH3JDP3G4SNENE4IOHQM7B","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":"3e297988184a8344ac5fd291573c6eee3436ceaee0cd1358d4673f22baeca238","cross_cats_sorted":["cs.MM"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-07T10:27:31Z","title_canon_sha256":"245e8a4451867a8ea4555ae2768891162eab1daa59297e4bef6bb4ba6c82e9f6"},"schema_version":"1.0","source":{"id":"2607.06118","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.06118","created_at":"2026-07-08T01:18:57Z"},{"alias_kind":"arxiv_version","alias_value":"2607.06118v1","created_at":"2026-07-08T01:18:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.06118","created_at":"2026-07-08T01:18:57Z"},{"alias_kind":"pith_short_12","alias_value":"3ZRBUH3JDP3G","created_at":"2026-07-08T01:18:57Z"},{"alias_kind":"pith_short_16","alias_value":"3ZRBUH3JDP3G4SNE","created_at":"2026-07-08T01:18:57Z"},{"alias_kind":"pith_short_8","alias_value":"3ZRBUH3J","created_at":"2026-07-08T01:18:57Z"}],"graph_snapshots":[{"event_id":"sha256:a96f842473793e2579001ca224afc865370fb8da44b77ae89d8b252c5d752b4c","target":"graph","created_at":"2026-07-08T01:18:57Z","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/2607.06118/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As web agents increasingly demonstrate capabilities in automated task execution, the development of robust evaluation frameworks for assessing their navigation and task completion performance has emerged as a critical research priority. However, existing benchmarks exhibit fundamental limitations. First, they suffer from insufficient scale and limited domain diversity, constraining comprehensive evaluation of cross-domain generalization. Second, prevailing LLM-as-Judge evaluation methodologies inadequately capture fine-grained interaction semantics, particularly regarding precise query formula","authors_text":"Anyang Su, Chenxu Zhao, Hanning Wang, Minghui Wu, Ping Jiang, Shuo Wang, Tianyu Fu, Wei Dong, Yuyang Chen, Zhen Lei, Zhe Yu, Zhizhou Fang","cross_cats":["cs.MM"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-07T10:27:31Z","title":"WebRetriever: A Large-Scale Comprehensive Benchmark for Efficient Web Agent Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.06118","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":"sha256:75e0e018f3214629298fd0d6e2049715d678bac8e50a92ef1444655182e65a2b","target":"record","created_at":"2026-07-08T01:18:57Z","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":"3e297988184a8344ac5fd291573c6eee3436ceaee0cd1358d4673f22baeca238","cross_cats_sorted":["cs.MM"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-07T10:27:31Z","title_canon_sha256":"245e8a4451867a8ea4555ae2768891162eab1daa59297e4bef6bb4ba6c82e9f6"},"schema_version":"1.0","source":{"id":"2607.06118","kind":"arxiv","version":1}},"canonical_sha256":"de621a1f691bf66e49a46938871e0cf878446db0ebba83eb5913110b6ad1dd90","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"de621a1f691bf66e49a46938871e0cf878446db0ebba83eb5913110b6ad1dd90","first_computed_at":"2026-07-08T01:18:57.413939Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-08T01:18:57.413939Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"cexW5WCChc5f7TvrnaIbSZKEgaoUl212fPG4BbVGYunpV/Jw4F+GOGYyItSxoG73nnmvGMo4+QzSOTNPheVSCQ==","signature_status":"signed_v1","signed_at":"2026-07-08T01:18:57.414432Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.06118","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:75e0e018f3214629298fd0d6e2049715d678bac8e50a92ef1444655182e65a2b","sha256:a96f842473793e2579001ca224afc865370fb8da44b77ae89d8b252c5d752b4c"],"state_sha256":"96037f9b5f8929ef2f2d309fccd97df96b9ac4976da20f57ce28e51dfad85383"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"EXsZgZZoNbEHU53fLnQMZ7al+9lXUt+R5nDzKaXzi8UYFMXw/hc2SE2nxz2u9O63PbHLlp2+87/+i7fJZ0t2DQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-11T14:42:44.316347Z","bundle_sha256":"48dc191238dc10cc1ed7424ba535c885b3b0b6e917c03f4706ec18ce82800d70"}}