{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:6TUDGJIOV6IOP3XZBDOUHGZHGN","short_pith_number":"pith:6TUDGJIO","schema_version":"1.0","canonical_sha256":"f4e833250eaf90e7eef908dd439b273363f432249a5c773e36f5ceb986501bf1","source":{"kind":"arxiv","id":"2605.27134","version":1},"attestation_state":"computed","paper":{"title":"Scaling, Benchmarking, and Reasoning of Vision-Language Agents for Mobile GUI Navigation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Heng Qu, Jian Luan, Pengzhi Gao, Renren Jin, Wei Liu, Wenzong Zhang, Yike Liu","submitted_at":"2026-05-26T15:03:56Z","abstract_excerpt":"Vision-Language Models (VLMs) have shown rapid progress in mobile GUI navigation. This paper presents a systematic study of data scaling, benchmarking, and reasoning for VLM-based agents in this domain. To facilitate rigorous evaluation, we introduce HyperTrack, a large-scale dataset with over 16000 real-world tasks across more than 650 Chinese mobile applications, along with GUIEvalKit, an open-source toolkit for unified benchmarking of VLMs on offline GUI navigation tasks. Using HyperTrack, we analyze the effects of training data scale on both supervised and reinforcement-based finetuning. O"},"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":"2605.27134","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-26T15:03:56Z","cross_cats_sorted":[],"title_canon_sha256":"c05c8519896601efb855e3670cdbdc7945bbffa36fc18968bb1af7a620bd9ded","abstract_canon_sha256":"6cbb4e0e971b35b1edbbe580fa5e3e697f1f41e4ff5972f2a8fa68deae1b22e2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-27T02:05:43.633766Z","signature_b64":"CIZz66rhvrLhn9DfZ6q7iwXtMhjeHHGCUkxMJS460oOCSFdsE1IAHDUXKR7+BRgyMTziCTFkB31vw3UX7SdvBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f4e833250eaf90e7eef908dd439b273363f432249a5c773e36f5ceb986501bf1","last_reissued_at":"2026-05-27T02:05:43.632916Z","signature_status":"signed_v1","first_computed_at":"2026-05-27T02:05:43.632916Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scaling, Benchmarking, and Reasoning of Vision-Language Agents for Mobile GUI Navigation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Heng Qu, Jian Luan, Pengzhi Gao, Renren Jin, Wei Liu, Wenzong Zhang, Yike Liu","submitted_at":"2026-05-26T15:03:56Z","abstract_excerpt":"Vision-Language Models (VLMs) have shown rapid progress in mobile GUI navigation. This paper presents a systematic study of data scaling, benchmarking, and reasoning for VLM-based agents in this domain. To facilitate rigorous evaluation, we introduce HyperTrack, a large-scale dataset with over 16000 real-world tasks across more than 650 Chinese mobile applications, along with GUIEvalKit, an open-source toolkit for unified benchmarking of VLMs on offline GUI navigation tasks. Using HyperTrack, we analyze the effects of training data scale on both supervised and reinforcement-based finetuning. O"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.27134","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/2605.27134/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":"2605.27134","created_at":"2026-05-27T02:05:43.633053+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.27134v1","created_at":"2026-05-27T02:05:43.633053+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.27134","created_at":"2026-05-27T02:05:43.633053+00:00"},{"alias_kind":"pith_short_12","alias_value":"6TUDGJIOV6IO","created_at":"2026-05-27T02:05:43.633053+00:00"},{"alias_kind":"pith_short_16","alias_value":"6TUDGJIOV6IOP3XZ","created_at":"2026-05-27T02:05:43.633053+00:00"},{"alias_kind":"pith_short_8","alias_value":"6TUDGJIO","created_at":"2026-05-27T02:05:43.633053+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/6TUDGJIOV6IOP3XZBDOUHGZHGN","json":"https://pith.science/pith/6TUDGJIOV6IOP3XZBDOUHGZHGN.json","graph_json":"https://pith.science/api/pith-number/6TUDGJIOV6IOP3XZBDOUHGZHGN/graph.json","events_json":"https://pith.science/api/pith-number/6TUDGJIOV6IOP3XZBDOUHGZHGN/events.json","paper":"https://pith.science/paper/6TUDGJIO"},"agent_actions":{"view_html":"https://pith.science/pith/6TUDGJIOV6IOP3XZBDOUHGZHGN","download_json":"https://pith.science/pith/6TUDGJIOV6IOP3XZBDOUHGZHGN.json","view_paper":"https://pith.science/paper/6TUDGJIO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.27134&json=true","fetch_graph":"https://pith.science/api/pith-number/6TUDGJIOV6IOP3XZBDOUHGZHGN/graph.json","fetch_events":"https://pith.science/api/pith-number/6TUDGJIOV6IOP3XZBDOUHGZHGN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6TUDGJIOV6IOP3XZBDOUHGZHGN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6TUDGJIOV6IOP3XZBDOUHGZHGN/action/storage_attestation","attest_author":"https://pith.science/pith/6TUDGJIOV6IOP3XZBDOUHGZHGN/action/author_attestation","sign_citation":"https://pith.science/pith/6TUDGJIOV6IOP3XZBDOUHGZHGN/action/citation_signature","submit_replication":"https://pith.science/pith/6TUDGJIOV6IOP3XZBDOUHGZHGN/action/replication_record"}},"created_at":"2026-05-27T02:05:43.633053+00:00","updated_at":"2026-05-27T02:05:43.633053+00:00"}