{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YUGAO2NTYSS7GZZNTSZQ6VZUBG","short_pith_number":"pith:YUGAO2NT","schema_version":"1.0","canonical_sha256":"c50c0769b3c4a5f3672d9cb30f5734099c1a9c9e3dfb291c1b1e64c3dc359de3","source":{"kind":"arxiv","id":"2508.20018","version":1},"attestation_state":"computed","paper":{"title":"SWIRL: A Staged Workflow for Interleaved Reinforcement Learning in Mobile GUI Control","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.CV","cs.MA"],"primary_cat":"cs.AI","authors_text":"Dahai Yu, Jin Wang, Michael K. Ng, Ping Luo, Quanfeng Lu, Shuai Zhong, Zhantao Ma","submitted_at":"2025-08-27T16:27:19Z","abstract_excerpt":"The rapid advancement of large vision language models (LVLMs) and agent systems has heightened interest in mobile GUI agents that can reliably translate natural language into interface operations. Existing single-agent approaches, however, remain limited by structural constraints. Although multi-agent systems naturally decouple different competencies, recent progress in multi-agent reinforcement learning (MARL) has often been hindered by inefficiency and remains incompatible with current LVLM architectures. To address these challenges, we introduce SWIRL, a staged workflow for interleaved rein"},"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":"2508.20018","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-08-27T16:27:19Z","cross_cats_sorted":["cs.CL","cs.CV","cs.MA"],"title_canon_sha256":"94cc0761e8d44b26dd647ca74f4f5d72ff45e3871ad13414858cd3146127a851","abstract_canon_sha256":"0c4de17a20cd62d39a4f9a7389f808650edd9b3db1ae034f0b865003a0b598af"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:00:22.257586Z","signature_b64":"YjptsujL/BW8JSAHHfXoYGIiAylQ2OQpTKoWbh1taHDG7Kz/XRGW4uBipwZEdG1554e+TGDa0L2/5f7WxJ51Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c50c0769b3c4a5f3672d9cb30f5734099c1a9c9e3dfb291c1b1e64c3dc359de3","last_reissued_at":"2026-07-05T12:00:22.257065Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:00:22.257065Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SWIRL: A Staged Workflow for Interleaved Reinforcement Learning in Mobile GUI Control","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.CV","cs.MA"],"primary_cat":"cs.AI","authors_text":"Dahai Yu, Jin Wang, Michael K. Ng, Ping Luo, Quanfeng Lu, Shuai Zhong, Zhantao Ma","submitted_at":"2025-08-27T16:27:19Z","abstract_excerpt":"The rapid advancement of large vision language models (LVLMs) and agent systems has heightened interest in mobile GUI agents that can reliably translate natural language into interface operations. Existing single-agent approaches, however, remain limited by structural constraints. Although multi-agent systems naturally decouple different competencies, recent progress in multi-agent reinforcement learning (MARL) has often been hindered by inefficiency and remains incompatible with current LVLM architectures. To address these challenges, we introduce SWIRL, a staged workflow for interleaved rein"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.20018","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/2508.20018/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":"2508.20018","created_at":"2026-07-05T12:00:22.257131+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.20018v1","created_at":"2026-07-05T12:00:22.257131+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.20018","created_at":"2026-07-05T12:00:22.257131+00:00"},{"alias_kind":"pith_short_12","alias_value":"YUGAO2NTYSS7","created_at":"2026-07-05T12:00:22.257131+00:00"},{"alias_kind":"pith_short_16","alias_value":"YUGAO2NTYSS7GZZN","created_at":"2026-07-05T12:00:22.257131+00:00"},{"alias_kind":"pith_short_8","alias_value":"YUGAO2NT","created_at":"2026-07-05T12:00:22.257131+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01249","citing_title":"Trust Region On-Policy Distillation","ref_index":30,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YUGAO2NTYSS7GZZNTSZQ6VZUBG","json":"https://pith.science/pith/YUGAO2NTYSS7GZZNTSZQ6VZUBG.json","graph_json":"https://pith.science/api/pith-number/YUGAO2NTYSS7GZZNTSZQ6VZUBG/graph.json","events_json":"https://pith.science/api/pith-number/YUGAO2NTYSS7GZZNTSZQ6VZUBG/events.json","paper":"https://pith.science/paper/YUGAO2NT"},"agent_actions":{"view_html":"https://pith.science/pith/YUGAO2NTYSS7GZZNTSZQ6VZUBG","download_json":"https://pith.science/pith/YUGAO2NTYSS7GZZNTSZQ6VZUBG.json","view_paper":"https://pith.science/paper/YUGAO2NT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.20018&json=true","fetch_graph":"https://pith.science/api/pith-number/YUGAO2NTYSS7GZZNTSZQ6VZUBG/graph.json","fetch_events":"https://pith.science/api/pith-number/YUGAO2NTYSS7GZZNTSZQ6VZUBG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YUGAO2NTYSS7GZZNTSZQ6VZUBG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YUGAO2NTYSS7GZZNTSZQ6VZUBG/action/storage_attestation","attest_author":"https://pith.science/pith/YUGAO2NTYSS7GZZNTSZQ6VZUBG/action/author_attestation","sign_citation":"https://pith.science/pith/YUGAO2NTYSS7GZZNTSZQ6VZUBG/action/citation_signature","submit_replication":"https://pith.science/pith/YUGAO2NTYSS7GZZNTSZQ6VZUBG/action/replication_record"}},"created_at":"2026-07-05T12:00:22.257131+00:00","updated_at":"2026-07-05T12:00:22.257131+00:00"}