{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:26GH5WORHYUEK372LHSES7T333","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"1505f23c909af627f22c2459b06139f2a518910bb0d28529294ed56f6b4fb403","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-15T02:27:41Z","title_canon_sha256":"72b5086792ce21d0a8928bc06558813c669fe1b879e949a1205315e775d4f736"},"schema_version":"1.0","source":{"id":"2605.15542","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.15542","created_at":"2026-05-20T00:01:04Z"},{"alias_kind":"arxiv_version","alias_value":"2605.15542v1","created_at":"2026-05-20T00:01:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.15542","created_at":"2026-05-20T00:01:04Z"},{"alias_kind":"pith_short_12","alias_value":"26GH5WORHYUE","created_at":"2026-05-20T00:01:04Z"},{"alias_kind":"pith_short_16","alias_value":"26GH5WORHYUEK372","created_at":"2026-05-20T00:01:04Z"},{"alias_kind":"pith_short_8","alias_value":"26GH5WOR","created_at":"2026-05-20T00:01:04Z"}],"graph_snapshots":[{"event_id":"sha256:c53ba1e070292c4454b53f604965cf00db5e2e368f1b1a0ea952cfc583341b43","target":"graph","created_at":"2026-05-20T00:01:04Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Experiments demonstrate that DRS-GUI yields a 14% improvement on ScreenSpot-Pro for general and GUI-specific MLLMs (Qwen2.5-VL-7B and UGround-V1-7B), significantly enhancing grounding performance and generalization."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The three perceptual actions (Focus, Shift, and Scatter) performed by the lightweight UI Perceptor, when scheduled by the MCTS-based Action Planner, will reliably generate and select instruction-relevant region proposals from cluttered high-resolution screenshots without requiring any model training or fine-tuning."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"DRS-GUI introduces a dynamic region search method with Focus/Shift/Scatter actions and MCTS-based planning that improves GUI grounding accuracy by 14% on ScreenSpot-Pro for both general and GUI-specific MLLMs without any training."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A training-free dynamic region search method improves GUI grounding performance by 14 percent in existing multimodal models."}],"snapshot_sha256":"c61e82e9070dc21b7ac5c50d7dd27a591462b53c6dbe2a0079a320ed72ed6c66"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"1d3f1858a8a791bb625e3df1449a4e969a8d75fd2c50c6dacc74869c4075ee2a"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T14:31:17.431374Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T14:27:26.780987Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"cited_work_retraction","ran_at":"2026-05-19T14:22:00.981090Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-19T14:21:54.025480Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"shingle_duplication","ran_at":"2026-05-19T13:49:41.825869Z","status":"skipped","version":"0.1.0"},{"findings_count":0,"name":"citation_quote_validity","ran_at":"2026-05-19T13:49:41.363442Z","status":"skipped","version":"0.1.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T13:33:22.611102Z","status":"skipped","version":"1.0.0"}],"endpoint":"/pith/2605.15542/integrity.json","findings":[],"snapshot_sha256":"190d8ee564ffea5cfd7a5b068ab5e849c38054971c89f8f1f943d7f3302ab476","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"GUI agents powered by Multimodal Large Language Models (MLLMs) have demonstrated impressive capability in understanding and executing user instructions. However, accurately grounding instruction-relevant elements from high-resolution screenshots cluttered with irrelevant UI components remains challenging for existing approaches. Inspired by how humans dynamically adjust their perceptual scope to locate task-related regions on complex screens, we propose DRS-GUI, a training-free dynamic region search framework for GUI grounding that can be seamlessly integrated into existing MLLMs. DRS-GUI intr","authors_text":"Huawen Shen, Liu Yu, Shiyu Liu, Yichao Liu, Yu Zhou, Zeyu Chen","cross_cats":[],"headline":"A training-free dynamic region search method improves GUI grounding performance by 14 percent in existing multimodal models.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-15T02:27:41Z","title":"DRS-GUI: Dynamic Region Search for Training-Free GUI Grounding"},"references":{"count":54,"internal_anchors":17,"resolved_work":54,"sample":[{"cited_arxiv_id":"2303.08774","doi":"","is_internal_anchor":true,"ref_index":1,"title":"GPT-4 Technical Report","work_id":"b928e041-6991-4c08-8c81-0359e4097c7b","year":null},{"cited_arxiv_id":"2309.16609","doi":"","is_internal_anchor":true,"ref_index":2,"title":"Qwen Technical Report","work_id":"bb1fd52f-6b2f-437c-9516-37bdf6eb9be8","year":2023},{"cited_arxiv_id":"2308.12966","doi":"","is_internal_anchor":true,"ref_index":3,"title":"Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond","work_id":"cbc2bb21-b6bb-46c0-80bf-107e195ffe10","year":2023},{"cited_arxiv_id":"2502.13923","doi":"","is_internal_anchor":true,"ref_index":4,"title":"Qwen2.5-VL Technical Report","work_id":"69dffacb-bfe8-442d-be86-48624c60426f","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Introducing our multimodal models, 2023","work_id":"5e64299f-ba20-4018-9460-81ed6cca321b","year":2023}],"snapshot_sha256":"85d9a02f82f3edb09c8ed74c6989a51f357399345983fbeca897831fd43cd0cc"},"source":{"id":"2605.15542","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-19T14:20:03.742235Z","id":"7dc9aa6b-3082-4e00-9a7e-ffab5cf2943a","model_set":{"reader":"grok-4.3"},"one_line_summary":"DRS-GUI introduces a dynamic region search method with Focus/Shift/Scatter actions and MCTS-based planning that improves GUI grounding accuracy by 14% on ScreenSpot-Pro for both general and GUI-specific MLLMs without any training.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A training-free dynamic region search method improves GUI grounding performance by 14 percent in existing multimodal models.","strongest_claim":"Experiments demonstrate that DRS-GUI yields a 14% improvement on ScreenSpot-Pro for general and GUI-specific MLLMs (Qwen2.5-VL-7B and UGround-V1-7B), significantly enhancing grounding performance and generalization.","weakest_assumption":"The three perceptual actions (Focus, Shift, and Scatter) performed by the lightweight UI Perceptor, when scheduled by the MCTS-based Action Planner, will reliably generate and select instruction-relevant region proposals from cluttered high-resolution screenshots without requiring any model training or fine-tuning."}},"verdict_id":"7dc9aa6b-3082-4e00-9a7e-ffab5cf2943a"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:c5a2c632377c6b95441dfe9937610a1be68112f9ff3afd7d5444d5424de578c0","target":"record","created_at":"2026-05-20T00:01:04Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"1505f23c909af627f22c2459b06139f2a518910bb0d28529294ed56f6b4fb403","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-15T02:27:41Z","title_canon_sha256":"72b5086792ce21d0a8928bc06558813c669fe1b879e949a1205315e775d4f736"},"schema_version":"1.0","source":{"id":"2605.15542","kind":"arxiv","version":1}},"canonical_sha256":"d78c7ed9d13e28456ffa59e4497e7bdeeeaf92c0a6f92aebf01b29f637e2897c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d78c7ed9d13e28456ffa59e4497e7bdeeeaf92c0a6f92aebf01b29f637e2897c","first_computed_at":"2026-05-20T00:01:04.400686Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:01:04.400686Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Ly4dmNMnQ3gFONMKI9ZR5eik/CwfII3B+hhcBBSWyg51K1zK3YitPA3Gqv2+/0mUYWMQs91YAwQ/E09Zfjw9Dw==","signature_status":"signed_v1","signed_at":"2026-05-20T00:01:04.401361Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.15542","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c5a2c632377c6b95441dfe9937610a1be68112f9ff3afd7d5444d5424de578c0","sha256:c53ba1e070292c4454b53f604965cf00db5e2e368f1b1a0ea952cfc583341b43"],"state_sha256":"6a69adc8d3110ed776c330ada00047d41ce2da46f40c9272244463c94d3d1eb6"}