{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:QS2YWBRC2BXEGVWADUHSHR2DNE","short_pith_number":"pith:QS2YWBRC","schema_version":"1.0","canonical_sha256":"84b58b0622d06e4356c01d0f23c74369040c31ac65084a2aa821404f93c1c310","source":{"kind":"arxiv","id":"2502.02955","version":1},"attestation_state":"computed","paper":{"title":"ReachAgent: Enhancing Mobile Agent via Page Reaching and Operation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bin Wang, Jian Luan, Qinzhuo Wu, Wei Liu","submitted_at":"2025-02-05T07:35:23Z","abstract_excerpt":"Recently, mobile AI agents have gained increasing attention. Given a task, mobile AI agents can interact with mobile devices in multiple steps and finally form a GUI flow that solves the task. However, existing agents tend to focus on most task-relevant elements at each step, leading to local optimal solutions and ignoring the overall GUI flow. To address this issue, we constructed a training dataset called MobileReach, which breaks the task into page reaching and operation subtasks. Furthermore, we propose ReachAgent, a two-stage framework that focuses on improving its task-completion abiliti"},"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":"2502.02955","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-05T07:35:23Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c6c0b843366924fa055bfc5cd6347eb79a92e5940176a2c522c229cde3a9e785","abstract_canon_sha256":"c5042a23fd4aa541f202a7cc5b16931d349b7fdb16d18c035cded0326a4b5fb6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:09:57.456262Z","signature_b64":"8RN/Iy4U6t6Y3Z5cFRZIdD0xz5qJgDkmsW4hHR/4wjBUrzFd/GWtj9ibP6Dbq129hQPS+1I0f1Hpdnostt1KDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"84b58b0622d06e4356c01d0f23c74369040c31ac65084a2aa821404f93c1c310","last_reissued_at":"2026-07-05T10:09:57.455752Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:09:57.455752Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ReachAgent: Enhancing Mobile Agent via Page Reaching and Operation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bin Wang, Jian Luan, Qinzhuo Wu, Wei Liu","submitted_at":"2025-02-05T07:35:23Z","abstract_excerpt":"Recently, mobile AI agents have gained increasing attention. Given a task, mobile AI agents can interact with mobile devices in multiple steps and finally form a GUI flow that solves the task. However, existing agents tend to focus on most task-relevant elements at each step, leading to local optimal solutions and ignoring the overall GUI flow. To address this issue, we constructed a training dataset called MobileReach, which breaks the task into page reaching and operation subtasks. Furthermore, we propose ReachAgent, a two-stage framework that focuses on improving its task-completion abiliti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.02955","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/2502.02955/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":"2502.02955","created_at":"2026-07-05T10:09:57.455817+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.02955v1","created_at":"2026-07-05T10:09:57.455817+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.02955","created_at":"2026-07-05T10:09:57.455817+00:00"},{"alias_kind":"pith_short_12","alias_value":"QS2YWBRC2BXE","created_at":"2026-07-05T10:09:57.455817+00:00"},{"alias_kind":"pith_short_16","alias_value":"QS2YWBRC2BXEGVWA","created_at":"2026-07-05T10:09:57.455817+00:00"},{"alias_kind":"pith_short_8","alias_value":"QS2YWBRC","created_at":"2026-07-05T10:09:57.455817+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.21953","citing_title":"MapAgent: Trajectory-Constructed Memory-Augmented Planning for Mobile Task Automation","ref_index":41,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QS2YWBRC2BXEGVWADUHSHR2DNE","json":"https://pith.science/pith/QS2YWBRC2BXEGVWADUHSHR2DNE.json","graph_json":"https://pith.science/api/pith-number/QS2YWBRC2BXEGVWADUHSHR2DNE/graph.json","events_json":"https://pith.science/api/pith-number/QS2YWBRC2BXEGVWADUHSHR2DNE/events.json","paper":"https://pith.science/paper/QS2YWBRC"},"agent_actions":{"view_html":"https://pith.science/pith/QS2YWBRC2BXEGVWADUHSHR2DNE","download_json":"https://pith.science/pith/QS2YWBRC2BXEGVWADUHSHR2DNE.json","view_paper":"https://pith.science/paper/QS2YWBRC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.02955&json=true","fetch_graph":"https://pith.science/api/pith-number/QS2YWBRC2BXEGVWADUHSHR2DNE/graph.json","fetch_events":"https://pith.science/api/pith-number/QS2YWBRC2BXEGVWADUHSHR2DNE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QS2YWBRC2BXEGVWADUHSHR2DNE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QS2YWBRC2BXEGVWADUHSHR2DNE/action/storage_attestation","attest_author":"https://pith.science/pith/QS2YWBRC2BXEGVWADUHSHR2DNE/action/author_attestation","sign_citation":"https://pith.science/pith/QS2YWBRC2BXEGVWADUHSHR2DNE/action/citation_signature","submit_replication":"https://pith.science/pith/QS2YWBRC2BXEGVWADUHSHR2DNE/action/replication_record"}},"created_at":"2026-07-05T10:09:57.455817+00:00","updated_at":"2026-07-05T10:09:57.455817+00:00"}