{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:LAN5VRYUQT5HM65KPEPCELB2MB","short_pith_number":"pith:LAN5VRYU","schema_version":"1.0","canonical_sha256":"581bdac71484fa767baa791e222c3a606ec321fed3bfbaa56a89988590ed42e9","source":{"kind":"arxiv","id":"2503.07919","version":3},"attestation_state":"computed","paper":{"title":"BEARCUBS: A benchmark for computer-using web agents","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.AI","authors_text":"Chau Minh Pham, Katherine Thai, Mazin Nadaf, Mohit Iyyer, Yapei Chang, Yixiao Song","submitted_at":"2025-03-10T23:50:30Z","abstract_excerpt":"Modern web agents possess computer use abilities that allow them to interact with webpages by sending commands to a virtual keyboard and mouse. While such agents have considerable potential to assist human users with complex tasks, evaluating their capabilities in real-world settings poses a major challenge. To this end, we introduce BEARCUBS, a \"smallbut mighty\" benchmark of 111 information-seeking questions designed to evaluate a web agent's ability to search, browse, and identify factual information from the web. Unlike prior web agent benchmarks, solving BEARCUBS requires (1) accessing liv"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2503.07919","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2025-03-10T23:50:30Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"81bae4ef1a8f07830b7bd25284e47cebed3404cd463a650b36ec2b46dee47dde","abstract_canon_sha256":"99c74a2347006b255e71fef678f25778aaedb235b5ac95748f3749b014f76b58"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:42:27.048434Z","signature_b64":"bHwkpdcridjX+pv6pVibHvZPQ9Sw9FU2rnGnIyL1ujBWpw8BsBa0GSzg6rfBBCewk1S0FZslnNKfPz0YHd7yBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"581bdac71484fa767baa791e222c3a606ec321fed3bfbaa56a89988590ed42e9","last_reissued_at":"2026-07-05T11:42:27.047927Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:42:27.047927Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BEARCUBS: A benchmark for computer-using web agents","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.AI","authors_text":"Chau Minh Pham, Katherine Thai, Mazin Nadaf, Mohit Iyyer, Yapei Chang, Yixiao Song","submitted_at":"2025-03-10T23:50:30Z","abstract_excerpt":"Modern web agents possess computer use abilities that allow them to interact with webpages by sending commands to a virtual keyboard and mouse. While such agents have considerable potential to assist human users with complex tasks, evaluating their capabilities in real-world settings poses a major challenge. To this end, we introduce BEARCUBS, a \"smallbut mighty\" benchmark of 111 information-seeking questions designed to evaluate a web agent's ability to search, browse, and identify factual information from the web. Unlike prior web agent benchmarks, solving BEARCUBS requires (1) accessing liv"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.07919","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/2503.07919/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2503.07919","created_at":"2026-07-05T11:42:27.047984+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.07919v3","created_at":"2026-07-05T11:42:27.047984+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.07919","created_at":"2026-07-05T11:42:27.047984+00:00"},{"alias_kind":"pith_short_12","alias_value":"LAN5VRYUQT5H","created_at":"2026-07-05T11:42:27.047984+00:00"},{"alias_kind":"pith_short_16","alias_value":"LAN5VRYUQT5HM65K","created_at":"2026-07-05T11:42:27.047984+00:00"},{"alias_kind":"pith_short_8","alias_value":"LAN5VRYU","created_at":"2026-07-05T11:42:27.047984+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.06118","citing_title":"WebRetriever: A Large-Scale Comprehensive Benchmark for Efficient Web Agent Evaluation","ref_index":37,"is_internal_anchor":true},{"citing_arxiv_id":"2606.20683","citing_title":"From Question Answering to Task Completion: A Survey on Agent System and Harness Design","ref_index":88,"is_internal_anchor":false},{"citing_arxiv_id":"2601.18842","citing_title":"GUIGuard-Bench: Toward a General Evaluation for Privacy-Preserving GUI Agents","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2504.19678","citing_title":"From LLM Reasoning to Autonomous AI Agents: A Comprehensive Review","ref_index":117,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LAN5VRYUQT5HM65KPEPCELB2MB","json":"https://pith.science/pith/LAN5VRYUQT5HM65KPEPCELB2MB.json","graph_json":"https://pith.science/api/pith-number/LAN5VRYUQT5HM65KPEPCELB2MB/graph.json","events_json":"https://pith.science/api/pith-number/LAN5VRYUQT5HM65KPEPCELB2MB/events.json","paper":"https://pith.science/paper/LAN5VRYU"},"agent_actions":{"view_html":"https://pith.science/pith/LAN5VRYUQT5HM65KPEPCELB2MB","download_json":"https://pith.science/pith/LAN5VRYUQT5HM65KPEPCELB2MB.json","view_paper":"https://pith.science/paper/LAN5VRYU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.07919&json=true","fetch_graph":"https://pith.science/api/pith-number/LAN5VRYUQT5HM65KPEPCELB2MB/graph.json","fetch_events":"https://pith.science/api/pith-number/LAN5VRYUQT5HM65KPEPCELB2MB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LAN5VRYUQT5HM65KPEPCELB2MB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LAN5VRYUQT5HM65KPEPCELB2MB/action/storage_attestation","attest_author":"https://pith.science/pith/LAN5VRYUQT5HM65KPEPCELB2MB/action/author_attestation","sign_citation":"https://pith.science/pith/LAN5VRYUQT5HM65KPEPCELB2MB/action/citation_signature","submit_replication":"https://pith.science/pith/LAN5VRYUQT5HM65KPEPCELB2MB/action/replication_record"}},"created_at":"2026-07-05T11:42:27.047984+00:00","updated_at":"2026-07-05T11:42:27.047984+00:00"}