{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:SS3JRCD332SLBYB6NGMXQQZVU4","short_pith_number":"pith:SS3JRCD3","schema_version":"1.0","canonical_sha256":"94b698887bdea4b0e03e6999784335a711978c8009ca086c1c494069066dad7b","source":{"kind":"arxiv","id":"2412.04454","version":2},"attestation_state":"computed","paper":{"title":"Aguvis: Unified Pure Vision Agents for Autonomous GUI Interaction","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Amrita Saha, Caiming Xiong, Doyen Sahoo, Dunjie Lu, Junli Wang, Tao Yu, Tianbao Xie, Yiheng Xu, Zekun Wang","submitted_at":"2024-12-05T18:58:26Z","abstract_excerpt":"Automating GUI tasks remains challenging due to reliance on textual representations, platform-specific action spaces, and limited reasoning capabilities. We introduce Aguvis, a unified vision-based framework for autonomous GUI agents that directly operates on screen images, standardizes cross-platform interactions and incorporates structured reasoning via inner monologue. To enable this, we construct Aguvis Data Collection, a large-scale dataset with multimodal grounding and reasoning annotations, and develop a two-stage training pipeline that separates GUI grounding from planning and reasonin"},"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":"2412.04454","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-05T18:58:26Z","cross_cats_sorted":[],"title_canon_sha256":"a5deb504993ca2dabbfe454d87f27a0c5412959e16ffd63d95fdcb42d9d3c974","abstract_canon_sha256":"a49b0464fcb57a8d5ff4ed96178df17e73a14922a224929076a8fa6753c05eeb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T04:03:03.836585Z","signature_b64":"4fkDDXmSw4KxaLF8IBTpAzLFMPLLhuxCT8uPcBQY464/XSfjpx+A5yxIEQ52N3ma05SLjEZFhEMZTf92Ket3DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"94b698887bdea4b0e03e6999784335a711978c8009ca086c1c494069066dad7b","last_reissued_at":"2026-05-18T04:03:03.835929Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T04:03:03.835929Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Aguvis: Unified Pure Vision Agents for Autonomous GUI Interaction","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Amrita Saha, Caiming Xiong, Doyen Sahoo, Dunjie Lu, Junli Wang, Tao Yu, Tianbao Xie, Yiheng Xu, Zekun Wang","submitted_at":"2024-12-05T18:58:26Z","abstract_excerpt":"Automating GUI tasks remains challenging due to reliance on textual representations, platform-specific action spaces, and limited reasoning capabilities. We introduce Aguvis, a unified vision-based framework for autonomous GUI agents that directly operates on screen images, standardizes cross-platform interactions and incorporates structured reasoning via inner monologue. To enable this, we construct Aguvis Data Collection, a large-scale dataset with multimodal grounding and reasoning annotations, and develop a two-stage training pipeline that separates GUI grounding from planning and reasonin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.04454","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":""},"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":"2412.04454","created_at":"2026-05-18T04:03:03.836023+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.04454v2","created_at":"2026-05-18T04:03:03.836023+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.04454","created_at":"2026-05-18T04:03:03.836023+00:00"},{"alias_kind":"pith_short_12","alias_value":"SS3JRCD332SL","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"SS3JRCD332SLBYB6","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"SS3JRCD3","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":20,"internal_anchor_count":20,"sample":[{"citing_arxiv_id":"2509.21982","citing_title":"RISK: A Framework for GUI Agents in E-commerce Risk Management","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2510.10073","citing_title":"SecureWebArena: A Holistic Security Evaluation Benchmark for LVLM-based Web Agents","ref_index":51,"is_internal_anchor":true},{"citing_arxiv_id":"2510.24168","citing_title":"MGA: Memory-Driven GUI Agent for Observation-Centric Interaction","ref_index":36,"is_internal_anchor":true},{"citing_arxiv_id":"2507.05791","citing_title":"GTA1: GUI Test-time Scaling Agent","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2512.12634","citing_title":"MobiBench: Multi-Branch, Modular Benchmark for Mobile GUI Agents","ref_index":46,"is_internal_anchor":true},{"citing_arxiv_id":"2512.19396","citing_title":"EchoTrail-GUI: Building Actionable Memory for GUI Agents via Critic-Guided Self-Exploration","ref_index":32,"is_internal_anchor":true},{"citing_arxiv_id":"2503.21620","citing_title":"UI-R1: Enhancing Efficient Action Prediction of GUI Agents by Reinforcement Learning","ref_index":16,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14311","citing_title":"Beyond Binary: Reframing GUI Critique as Continuous Semantic Alignment","ref_index":120,"is_internal_anchor":true},{"citing_arxiv_id":"2509.02544","citing_title":"UI-TARS-2 Technical Report: Advancing GUI Agent with Multi-Turn Reinforcement Learning","ref_index":77,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12501","citing_title":"Covering Human Action Space for Computer Use: Data Synthesis and Benchmark","ref_index":31,"is_internal_anchor":true},{"citing_arxiv_id":"2604.25380","citing_title":"Benchmarking and Improving GUI Agents in High-Dynamic Environments","ref_index":40,"is_internal_anchor":true},{"citing_arxiv_id":"2604.22558","citing_title":"SOLAR-RL: Semi-Online Long-horizon Assignment Reinforcement Learning","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06664","citing_title":"BAMI: Training-Free Bias Mitigation in GUI Grounding","ref_index":33,"is_internal_anchor":true},{"citing_arxiv_id":"2604.21268","citing_title":"Measure Twice, Click Once: Co-evolving Proposer and Visual Critic via Reinforcement Learning for GUI Grounding","ref_index":63,"is_internal_anchor":true},{"citing_arxiv_id":"2604.06995","citing_title":"What's Missing in Screen-to-Action? Towards a UI-in-the-Loop Paradigm for Multimodal GUI Reasoning","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2604.25380","citing_title":"Benchmarking and Improving GUI Agents in High-Dynamic Environments","ref_index":40,"is_internal_anchor":true},{"citing_arxiv_id":"2604.14113","citing_title":"UI-Zoomer: Uncertainty-Driven Adaptive Zoom-In for GUI Grounding","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2604.13488","citing_title":"Towards Scalable Lightweight GUI Agents via Multi-role Orchestration","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2508.18265","citing_title":"InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency","ref_index":158,"is_internal_anchor":true},{"citing_arxiv_id":"2604.21375","citing_title":"VLAA-GUI: Knowing When to Stop, Recover, and Search, A Modular Framework for GUI Automation","ref_index":68,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SS3JRCD332SLBYB6NGMXQQZVU4","json":"https://pith.science/pith/SS3JRCD332SLBYB6NGMXQQZVU4.json","graph_json":"https://pith.science/api/pith-number/SS3JRCD332SLBYB6NGMXQQZVU4/graph.json","events_json":"https://pith.science/api/pith-number/SS3JRCD332SLBYB6NGMXQQZVU4/events.json","paper":"https://pith.science/paper/SS3JRCD3"},"agent_actions":{"view_html":"https://pith.science/pith/SS3JRCD332SLBYB6NGMXQQZVU4","download_json":"https://pith.science/pith/SS3JRCD332SLBYB6NGMXQQZVU4.json","view_paper":"https://pith.science/paper/SS3JRCD3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.04454&json=true","fetch_graph":"https://pith.science/api/pith-number/SS3JRCD332SLBYB6NGMXQQZVU4/graph.json","fetch_events":"https://pith.science/api/pith-number/SS3JRCD332SLBYB6NGMXQQZVU4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SS3JRCD332SLBYB6NGMXQQZVU4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SS3JRCD332SLBYB6NGMXQQZVU4/action/storage_attestation","attest_author":"https://pith.science/pith/SS3JRCD332SLBYB6NGMXQQZVU4/action/author_attestation","sign_citation":"https://pith.science/pith/SS3JRCD332SLBYB6NGMXQQZVU4/action/citation_signature","submit_replication":"https://pith.science/pith/SS3JRCD332SLBYB6NGMXQQZVU4/action/replication_record"}},"created_at":"2026-05-18T04:03:03.836023+00:00","updated_at":"2026-05-18T04:03:03.836023+00:00"}