{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:PEQSFWR4D2PDWVEO24ZOTI6QON","short_pith_number":"pith:PEQSFWR4","schema_version":"1.0","canonical_sha256":"792122da3c1e9e3b548ed732e9a3d0734b1dc4c397bd352fc46786d5954549a2","source":{"kind":"arxiv","id":"2605.29119","version":1},"attestation_state":"computed","paper":{"title":"PRO-CUA: Process-Reward Optimization for Computer Use Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Han Zhao, Hao Bai, Rui Yang, Tong Zhang, Yifei He","submitted_at":"2026-05-27T21:28:26Z","abstract_excerpt":"Computer use agents (CUAs) have shown strong potential for automating complex digital workflows, yet their training remains constrained by costly live environment interaction and limited high-quality supervision. Existing filtered behavior cloning pipelines suffer from imitation bottlenecks, including distribution shift from the expert demonstration and the absence of negative learning signals. Meanwhile, standard trajectory-level reinforcement learning struggles with sparse rewards, ambiguous credit assignment, and high infrastructure costs for long-horizon GUI interaction. In this work, we p"},"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":"2605.29119","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-27T21:28:26Z","cross_cats_sorted":[],"title_canon_sha256":"cb2a065b21808b4aa6faccb944a1388d8346eec97c98b7389a464bf7a85ec733","abstract_canon_sha256":"a035d74e69025c736500f3a1814d0084b09283924fd94897f099fe6a1ef1205a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:05:19.241795Z","signature_b64":"EUQeai/eUvYqwWmhUgeV+MFbZF7iIwxlMpdkt1yUrbBEwj3E54pQwEEYwLunvDV0nhAlQkWH168D9s48hZfbBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"792122da3c1e9e3b548ed732e9a3d0734b1dc4c397bd352fc46786d5954549a2","last_reissued_at":"2026-05-29T01:05:19.240854Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:05:19.240854Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PRO-CUA: Process-Reward Optimization for Computer Use Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Han Zhao, Hao Bai, Rui Yang, Tong Zhang, Yifei He","submitted_at":"2026-05-27T21:28:26Z","abstract_excerpt":"Computer use agents (CUAs) have shown strong potential for automating complex digital workflows, yet their training remains constrained by costly live environment interaction and limited high-quality supervision. Existing filtered behavior cloning pipelines suffer from imitation bottlenecks, including distribution shift from the expert demonstration and the absence of negative learning signals. Meanwhile, standard trajectory-level reinforcement learning struggles with sparse rewards, ambiguous credit assignment, and high infrastructure costs for long-horizon GUI interaction. In this work, we p"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.29119","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/2605.29119/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":"2605.29119","created_at":"2026-05-29T01:05:19.240972+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.29119v1","created_at":"2026-05-29T01:05:19.240972+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.29119","created_at":"2026-05-29T01:05:19.240972+00:00"},{"alias_kind":"pith_short_12","alias_value":"PEQSFWR4D2PD","created_at":"2026-05-29T01:05:19.240972+00:00"},{"alias_kind":"pith_short_16","alias_value":"PEQSFWR4D2PDWVEO","created_at":"2026-05-29T01:05:19.240972+00:00"},{"alias_kind":"pith_short_8","alias_value":"PEQSFWR4","created_at":"2026-05-29T01:05:19.240972+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/PEQSFWR4D2PDWVEO24ZOTI6QON","json":"https://pith.science/pith/PEQSFWR4D2PDWVEO24ZOTI6QON.json","graph_json":"https://pith.science/api/pith-number/PEQSFWR4D2PDWVEO24ZOTI6QON/graph.json","events_json":"https://pith.science/api/pith-number/PEQSFWR4D2PDWVEO24ZOTI6QON/events.json","paper":"https://pith.science/paper/PEQSFWR4"},"agent_actions":{"view_html":"https://pith.science/pith/PEQSFWR4D2PDWVEO24ZOTI6QON","download_json":"https://pith.science/pith/PEQSFWR4D2PDWVEO24ZOTI6QON.json","view_paper":"https://pith.science/paper/PEQSFWR4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.29119&json=true","fetch_graph":"https://pith.science/api/pith-number/PEQSFWR4D2PDWVEO24ZOTI6QON/graph.json","fetch_events":"https://pith.science/api/pith-number/PEQSFWR4D2PDWVEO24ZOTI6QON/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PEQSFWR4D2PDWVEO24ZOTI6QON/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PEQSFWR4D2PDWVEO24ZOTI6QON/action/storage_attestation","attest_author":"https://pith.science/pith/PEQSFWR4D2PDWVEO24ZOTI6QON/action/author_attestation","sign_citation":"https://pith.science/pith/PEQSFWR4D2PDWVEO24ZOTI6QON/action/citation_signature","submit_replication":"https://pith.science/pith/PEQSFWR4D2PDWVEO24ZOTI6QON/action/replication_record"}},"created_at":"2026-05-29T01:05:19.240972+00:00","updated_at":"2026-05-29T01:05:19.240972+00:00"}