{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:3SZUNEXQY3R56JMGXZ3DTUB4TX","short_pith_number":"pith:3SZUNEXQ","schema_version":"1.0","canonical_sha256":"dcb34692f0c6e3df2586be7639d03c9dc28cc81510c07e3b890648921b86b9ca","source":{"kind":"arxiv","id":"2607.23263","version":1},"attestation_state":"computed","paper":{"title":"SeekJudge: A Practical Reward Framework for Reinforcement Learning in Computer-Use Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Jierui Wang, Linchao Zhu, Yang Wan, Zhenhao Zhang","submitted_at":"2026-07-25T16:00:45Z","abstract_excerpt":"Deciding whether a trajectory actually fulfills its instruction governs how we measure computer-use agents on long-horizon graphical-user-interface tasks and how we train them with reinforcement learning. This judgment has long relied on rule-based evaluation, which struggles to align with human intention and goes stale when an app updates or its online content drifts. Existing model-based judges attempt to address these problems but still leave a performance gap to the rule-based evaluation. We propose the \\textbf{SeekJudge} framework, in which four role-specialized agents, a Condense, a Grou"},"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":"2607.23263","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-25T16:00:45Z","cross_cats_sorted":[],"title_canon_sha256":"0b20d28b2f8dd4aac82c1aee51b2a66fbd0ae804a4964be15465ddd3de69aa7c","abstract_canon_sha256":"b88627f3af26f6b5065f809623e270bbdafa55261e473d281bcb2073e8808827"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-28T01:22:44.657298Z","signature_b64":"hZBjkMtKNBtY3TfXrE5EthYsTs5OP6jMddQMCutZHpn5jSRxvTTDcvUVo2HGuZUN/4fSGyr63o6EdLAforEYBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dcb34692f0c6e3df2586be7639d03c9dc28cc81510c07e3b890648921b86b9ca","last_reissued_at":"2026-07-28T01:22:44.656432Z","signature_status":"signed_v1","first_computed_at":"2026-07-28T01:22:44.656432Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SeekJudge: A Practical Reward Framework for Reinforcement Learning in Computer-Use Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Jierui Wang, Linchao Zhu, Yang Wan, Zhenhao Zhang","submitted_at":"2026-07-25T16:00:45Z","abstract_excerpt":"Deciding whether a trajectory actually fulfills its instruction governs how we measure computer-use agents on long-horizon graphical-user-interface tasks and how we train them with reinforcement learning. This judgment has long relied on rule-based evaluation, which struggles to align with human intention and goes stale when an app updates or its online content drifts. Existing model-based judges attempt to address these problems but still leave a performance gap to the rule-based evaluation. We propose the \\textbf{SeekJudge} framework, in which four role-specialized agents, a Condense, a Grou"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.23263","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/2607.23263/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":"2607.23263","created_at":"2026-07-28T01:22:44.656882+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.23263v1","created_at":"2026-07-28T01:22:44.656882+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.23263","created_at":"2026-07-28T01:22:44.656882+00:00"},{"alias_kind":"pith_short_12","alias_value":"3SZUNEXQY3R5","created_at":"2026-07-28T01:22:44.656882+00:00"},{"alias_kind":"pith_short_16","alias_value":"3SZUNEXQY3R56JMG","created_at":"2026-07-28T01:22:44.656882+00:00"},{"alias_kind":"pith_short_8","alias_value":"3SZUNEXQ","created_at":"2026-07-28T01:22:44.656882+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/3SZUNEXQY3R56JMGXZ3DTUB4TX","json":"https://pith.science/pith/3SZUNEXQY3R56JMGXZ3DTUB4TX.json","graph_json":"https://pith.science/api/pith-number/3SZUNEXQY3R56JMGXZ3DTUB4TX/graph.json","events_json":"https://pith.science/api/pith-number/3SZUNEXQY3R56JMGXZ3DTUB4TX/events.json","paper":"https://pith.science/paper/3SZUNEXQ"},"agent_actions":{"view_html":"https://pith.science/pith/3SZUNEXQY3R56JMGXZ3DTUB4TX","download_json":"https://pith.science/pith/3SZUNEXQY3R56JMGXZ3DTUB4TX.json","view_paper":"https://pith.science/paper/3SZUNEXQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.23263&json=true","fetch_graph":"https://pith.science/api/pith-number/3SZUNEXQY3R56JMGXZ3DTUB4TX/graph.json","fetch_events":"https://pith.science/api/pith-number/3SZUNEXQY3R56JMGXZ3DTUB4TX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3SZUNEXQY3R56JMGXZ3DTUB4TX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3SZUNEXQY3R56JMGXZ3DTUB4TX/action/storage_attestation","attest_author":"https://pith.science/pith/3SZUNEXQY3R56JMGXZ3DTUB4TX/action/author_attestation","sign_citation":"https://pith.science/pith/3SZUNEXQY3R56JMGXZ3DTUB4TX/action/citation_signature","submit_replication":"https://pith.science/pith/3SZUNEXQY3R56JMGXZ3DTUB4TX/action/replication_record"}},"created_at":"2026-07-28T01:22:44.656882+00:00","updated_at":"2026-07-28T01:22:44.656882+00:00"}