{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:VPHUVQIWNZLB3LAGRAZPNLI6OE","short_pith_number":"pith:VPHUVQIW","schema_version":"1.0","canonical_sha256":"abcf4ac1166e561dac068832f6ad1e71034a49df4d32a1f97ea6b40a9693fca1","source":{"kind":"arxiv","id":"2505.12370","version":2},"attestation_state":"computed","paper":{"title":"Enhancing Visual Grounding for GUI Agents via Self-Evolutionary Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bo Li, Enguang Wang, Jian Zhang, Jie Chen, Jinwei Chen, Kaixin Li, Lujian Yao, Peng-Tao Jiang, Qibin Hou, Xinbin Yuan, Zhuoxuan Cai","submitted_at":"2025-05-18T11:22:04Z","abstract_excerpt":"Graphical User Interface (GUI) agents have made substantial strides in understanding and executing user instructions across diverse platforms. Yet, grounding these instructions to precise interface elements remains challenging, especially in complex, high-resolution, professional environments. Traditional supervised finetuning (SFT) methods often require large volumes of diverse data and exhibit weak generalization. To overcome these limitations, we introduce a reinforcement learning (RL) based framework that incorporates three core strategies: (1) seed data curation to ensure high quality tra"},"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":"2505.12370","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-18T11:22:04Z","cross_cats_sorted":[],"title_canon_sha256":"cb31d805dd6b4d4fb438e186137f2b4ad0473d1066e586952906e0a9c09073bd","abstract_canon_sha256":"be84a87f998bff2392cb11e102612063669f387c8d48d013f14626677710cb3c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:08:41.392694Z","signature_b64":"xCo8NBoyVkF/TqDyH1cYgs7r+/Yp9DyCxmhLo1mONgEsrY6dUPVCPcwA+tH/k7mmPkjLr1uAFeMkvYiR5vyhAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"abcf4ac1166e561dac068832f6ad1e71034a49df4d32a1f97ea6b40a9693fca1","last_reissued_at":"2026-07-05T11:08:41.392227Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:08:41.392227Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Enhancing Visual Grounding for GUI Agents via Self-Evolutionary Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bo Li, Enguang Wang, Jian Zhang, Jie Chen, Jinwei Chen, Kaixin Li, Lujian Yao, Peng-Tao Jiang, Qibin Hou, Xinbin Yuan, Zhuoxuan Cai","submitted_at":"2025-05-18T11:22:04Z","abstract_excerpt":"Graphical User Interface (GUI) agents have made substantial strides in understanding and executing user instructions across diverse platforms. Yet, grounding these instructions to precise interface elements remains challenging, especially in complex, high-resolution, professional environments. Traditional supervised finetuning (SFT) methods often require large volumes of diverse data and exhibit weak generalization. To overcome these limitations, we introduce a reinforcement learning (RL) based framework that incorporates three core strategies: (1) seed data curation to ensure high quality tra"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.12370","kind":"arxiv","version":2},"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/2505.12370/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":"2505.12370","created_at":"2026-07-05T11:08:41.392288+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.12370v2","created_at":"2026-07-05T11:08:41.392288+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.12370","created_at":"2026-07-05T11:08:41.392288+00:00"},{"alias_kind":"pith_short_12","alias_value":"VPHUVQIWNZLB","created_at":"2026-07-05T11:08:41.392288+00:00"},{"alias_kind":"pith_short_16","alias_value":"VPHUVQIWNZLB3LAG","created_at":"2026-07-05T11:08:41.392288+00:00"},{"alias_kind":"pith_short_8","alias_value":"VPHUVQIW","created_at":"2026-07-05T11:08:41.392288+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":23,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.13316","citing_title":"ReSum: Synergizing LLM Reasoning and Summarization with Reinforcement Learning","ref_index":69,"is_internal_anchor":false},{"citing_arxiv_id":"2606.10522","citing_title":"GUI-AC: Enhancing Continual Learning in GUI Agents","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29705","citing_title":"GUICrafter: Weakly-Supervised GUI Agent Leveraging Massive Unannotated Screenshots","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30884","citing_title":"GUI-C$^2$: Coarse-to-Fine GUI Grounding via Difficulty-Aware Reinforcement Learning","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16883","citing_title":"SE-GA: Memory-Augmented Self-Evolution for GUI Agents","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15542","citing_title":"DRS-GUI: Dynamic Region Search for Training-Free GUI Grounding","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2509.21982","citing_title":"RISK: A Framework for GUI Agents in E-commerce Risk Management","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2507.05791","citing_title":"GTA1: GUI Test-time Scaling Agent","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12549","citing_title":"What Happens Before Decoding? Prefill Determines GUI Grounding in VLMs","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2602.12430","citing_title":"Agent Skills for Large Language Models: Architecture, Acquisition, Security, and the Path Forward","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00642","citing_title":"Learn where to Click from Yourself: On-Policy Self-Distillation for GUI Grounding","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2604.25380","citing_title":"Benchmarking and Improving GUI Agents in High-Dynamic Environments","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2604.24348","citing_title":"OS-SPEAR: A Toolkit for the Safety, Performance,Efficiency, and Robustness Analysis of OS Agents","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06664","citing_title":"BAMI: Training-Free Bias Mitigation in GUI Grounding","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01208","citing_title":"Faithful Mobile GUI Agents with Guided Advantage Estimator","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00642","citing_title":"Learn where to Click from Yourself: On-Policy Self-Distillation for GUI Grounding","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2604.25380","citing_title":"Benchmarking and Improving GUI Agents in High-Dynamic Environments","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07110","citing_title":"Securing Computer-Use Agents: A Unified Architecture-Lifecycle Framework for Deployment-Grounded Reliability","ref_index":54,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07505","citing_title":"LiteGUI: Distilling Compact GUI Agents with Reinforcement Learning","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2604.07831","citing_title":"Are GUI Agents Focused Enough? Automated Distraction via Semantic-level UI Element Injection","ref_index":55,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14113","citing_title":"UI-Zoomer: Uncertainty-Driven Adaptive Zoom-In for GUI Grounding","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15376","citing_title":"Zoom Consistency: A Free Confidence Signal in Multi-Step Visual Grounding Pipelines","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02630","citing_title":"AutoFocus: Uncertainty-Aware Active Visual Search for GUI Grounding","ref_index":45,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VPHUVQIWNZLB3LAGRAZPNLI6OE","json":"https://pith.science/pith/VPHUVQIWNZLB3LAGRAZPNLI6OE.json","graph_json":"https://pith.science/api/pith-number/VPHUVQIWNZLB3LAGRAZPNLI6OE/graph.json","events_json":"https://pith.science/api/pith-number/VPHUVQIWNZLB3LAGRAZPNLI6OE/events.json","paper":"https://pith.science/paper/VPHUVQIW"},"agent_actions":{"view_html":"https://pith.science/pith/VPHUVQIWNZLB3LAGRAZPNLI6OE","download_json":"https://pith.science/pith/VPHUVQIWNZLB3LAGRAZPNLI6OE.json","view_paper":"https://pith.science/paper/VPHUVQIW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.12370&json=true","fetch_graph":"https://pith.science/api/pith-number/VPHUVQIWNZLB3LAGRAZPNLI6OE/graph.json","fetch_events":"https://pith.science/api/pith-number/VPHUVQIWNZLB3LAGRAZPNLI6OE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VPHUVQIWNZLB3LAGRAZPNLI6OE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VPHUVQIWNZLB3LAGRAZPNLI6OE/action/storage_attestation","attest_author":"https://pith.science/pith/VPHUVQIWNZLB3LAGRAZPNLI6OE/action/author_attestation","sign_citation":"https://pith.science/pith/VPHUVQIWNZLB3LAGRAZPNLI6OE/action/citation_signature","submit_replication":"https://pith.science/pith/VPHUVQIWNZLB3LAGRAZPNLI6OE/action/replication_record"}},"created_at":"2026-07-05T11:08:41.392288+00:00","updated_at":"2026-07-05T11:08:41.392288+00:00"}