{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:W7PQGLYBJTEPUD2U5RMR7ZYZ7Y","short_pith_number":"pith:W7PQGLYB","schema_version":"1.0","canonical_sha256":"b7df032f014cc8fa0f54ec591fe719fe00556868bb35f5281225a8339ae28371","source":{"kind":"arxiv","id":"2511.07332","version":2},"attestation_state":"computed","paper":{"title":"Grounding Computer Use Agents on Human Demonstrations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Aarash Feizi, Adriana Romero-Soriano, Christopher Pal, David Vazquez, Johan Obando-Ceron, Juan A. Rodriguez, Kaixin Li, Kevin Qinghong Lin, Nicolas Chapados, Perouz Taslakian, Rabiul Awal, Reihaneh Rabbany, Sai Rajeswar, Shravan Nayak, Spandana Gella, Xiangru Jian, Xing Han L\\`u","submitted_at":"2025-11-10T17:35:21Z","abstract_excerpt":"Building reliable computer-use agents requires grounding: accurately connecting natural language instructions to the correct on-screen elements. While large datasets exist for web and mobile interactions, high-quality resources for desktop environments are limited. To address this gap, we introduce GroundCUA, a large-scale desktop grounding dataset built from expert human demonstrations. It covers 87 applications across 12 categories and includes 56K screenshots, with every on-screen element carefully annotated for a total of over 3.56M human-verified annotations. From these demonstrations, we"},"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":"2511.07332","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-11-10T17:35:21Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"8250cf921724c32c95225c525012454dc1deee511abe866b4150e9e5f260881d","abstract_canon_sha256":"f178758047a798eb47d0feb10af85c4407834fa6bf28b5382af7a60390c6942b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-11T01:09:21.510967Z","signature_b64":"Y1lNKi8q7PW3sIwq7EPRGnVXV66EweV8w4BLrVD3+HMKSeTXKo0VEqFfIxCc5gMFN4cQAa6j5T+2s6tNKXWRCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b7df032f014cc8fa0f54ec591fe719fe00556868bb35f5281225a8339ae28371","last_reissued_at":"2026-06-11T01:09:21.509876Z","signature_status":"signed_v1","first_computed_at":"2026-06-11T01:09:21.509876Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Grounding Computer Use Agents on Human Demonstrations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Aarash Feizi, Adriana Romero-Soriano, Christopher Pal, David Vazquez, Johan Obando-Ceron, Juan A. Rodriguez, Kaixin Li, Kevin Qinghong Lin, Nicolas Chapados, Perouz Taslakian, Rabiul Awal, Reihaneh Rabbany, Sai Rajeswar, Shravan Nayak, Spandana Gella, Xiangru Jian, Xing Han L\\`u","submitted_at":"2025-11-10T17:35:21Z","abstract_excerpt":"Building reliable computer-use agents requires grounding: accurately connecting natural language instructions to the correct on-screen elements. While large datasets exist for web and mobile interactions, high-quality resources for desktop environments are limited. To address this gap, we introduce GroundCUA, a large-scale desktop grounding dataset built from expert human demonstrations. It covers 87 applications across 12 categories and includes 56K screenshots, with every on-screen element carefully annotated for a total of over 3.56M human-verified annotations. From these demonstrations, we"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2511.07332","kind":"arxiv","version":2},"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/2511.07332/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":"2511.07332","created_at":"2026-06-11T01:09:21.510027+00:00"},{"alias_kind":"arxiv_version","alias_value":"2511.07332v2","created_at":"2026-06-11T01:09:21.510027+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2511.07332","created_at":"2026-06-11T01:09:21.510027+00:00"},{"alias_kind":"pith_short_12","alias_value":"W7PQGLYBJTEP","created_at":"2026-06-11T01:09:21.510027+00:00"},{"alias_kind":"pith_short_16","alias_value":"W7PQGLYBJTEPUD2U","created_at":"2026-06-11T01:09:21.510027+00:00"},{"alias_kind":"pith_short_8","alias_value":"W7PQGLYB","created_at":"2026-06-11T01:09:21.510027+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/W7PQGLYBJTEPUD2U5RMR7ZYZ7Y","json":"https://pith.science/pith/W7PQGLYBJTEPUD2U5RMR7ZYZ7Y.json","graph_json":"https://pith.science/api/pith-number/W7PQGLYBJTEPUD2U5RMR7ZYZ7Y/graph.json","events_json":"https://pith.science/api/pith-number/W7PQGLYBJTEPUD2U5RMR7ZYZ7Y/events.json","paper":"https://pith.science/paper/W7PQGLYB"},"agent_actions":{"view_html":"https://pith.science/pith/W7PQGLYBJTEPUD2U5RMR7ZYZ7Y","download_json":"https://pith.science/pith/W7PQGLYBJTEPUD2U5RMR7ZYZ7Y.json","view_paper":"https://pith.science/paper/W7PQGLYB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2511.07332&json=true","fetch_graph":"https://pith.science/api/pith-number/W7PQGLYBJTEPUD2U5RMR7ZYZ7Y/graph.json","fetch_events":"https://pith.science/api/pith-number/W7PQGLYBJTEPUD2U5RMR7ZYZ7Y/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/W7PQGLYBJTEPUD2U5RMR7ZYZ7Y/action/timestamp_anchor","attest_storage":"https://pith.science/pith/W7PQGLYBJTEPUD2U5RMR7ZYZ7Y/action/storage_attestation","attest_author":"https://pith.science/pith/W7PQGLYBJTEPUD2U5RMR7ZYZ7Y/action/author_attestation","sign_citation":"https://pith.science/pith/W7PQGLYBJTEPUD2U5RMR7ZYZ7Y/action/citation_signature","submit_replication":"https://pith.science/pith/W7PQGLYBJTEPUD2U5RMR7ZYZ7Y/action/replication_record"}},"created_at":"2026-06-11T01:09:21.510027+00:00","updated_at":"2026-06-11T01:09:21.510027+00:00"}