{"paper":{"title":"Guide, Think, Act: Interactive Embodied Reasoning in Vision-Language-Action Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"GTA-VLA lets users steer vision-language-action models with explicit spatial visual cues for better robot control.","cross_cats":["cs.CV"],"primary_cat":"cs.RO","authors_text":"Chuanxiu Liu, Jie Liu, Jinghang Li, Lei Zhang, Qing Jiang, Qing Lian, Tianming Zhang, Xiaoke Jiang, Yiran Ling","submitted_at":"2026-05-13T14:58:29Z","abstract_excerpt":"In this paper, we propose GTA-VLA(Guide, Think, Act), an interactive Vision-Language-Action (VLA) framework that enables spatially steerable embodied reasoning by allowing users to guide robot policies with explicit visual cues. Existing VLA models learn a direct \"Sense-to-Act\" mapping from multimodal observations to robot actions. While effective within the training distribution, such tightly coupled policies are brittle under out-of-domain (OOD) shifts and difficult to correct when failures occur. Although recent embodied Chain-of-Thought (CoT) approaches expose intermediate reasoning, they "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"On the in-domain SimplerEnv WidowX benchmark, our framework achieves a state-of-the-art 81.2% success rate. Under OOD visual shifts and spatial ambiguities, a single visual interaction substantially improves task success over existing methods.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That users will supply accurate, task-relevant spatial priors (points, boxes, traces) that the model can reliably integrate without introducing new errors or ambiguities.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"GTA-VLA conditions VLA models on user spatial priors to produce a unified spatial-visual chain-of-thought, reaching 81.2% success on SimplerEnv WidowX and improving performance under out-of-distribution shifts.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"GTA-VLA lets users steer vision-language-action models with explicit spatial visual cues for better robot control.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"eaace184c0009b135198d08d3f0b278033038f551a69e5f68779487dc984677a"},"source":{"id":"2605.13632","kind":"arxiv","version":1},"verdict":{"id":"2527139b-5c9b-4368-a56b-41cd91fd7d10","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T18:27:09.637078Z","strongest_claim":"On the in-domain SimplerEnv WidowX benchmark, our framework achieves a state-of-the-art 81.2% success rate. Under OOD visual shifts and spatial ambiguities, a single visual interaction substantially improves task success over existing methods.","one_line_summary":"GTA-VLA conditions VLA models on user spatial priors to produce a unified spatial-visual chain-of-thought, reaching 81.2% success on SimplerEnv WidowX and improving performance under out-of-distribution shifts.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That users will supply accurate, task-relevant spatial priors (points, boxes, traces) that the model can reliably integrate without introducing new errors or ambiguities.","pith_extraction_headline":"GTA-VLA lets users steer vision-language-action models with explicit spatial visual cues for better robot control."},"references":{"count":39,"sample":[{"doi":"","year":2025,"title":"Qwen3-VL Technical Report","work_id":"1fe243aa-e3c0-4da6-b391-4cbcfc88d5c0","ref_index":1,"cited_arxiv_id":"2511.21631","is_internal_anchor":true},{"doi":"","year":2024,"title":"RT-H: Action Hierarchies Using Language","work_id":"ecf7cf18-c1a8-4a6b-bc2a-fb165643aa0d","ref_index":2,"cited_arxiv_id":"2403.01823","is_internal_anchor":true},{"doi":"","year":2024,"title":"$\\pi_0$: A Vision-Language-Action Flow Model for General Robot Control","work_id":"f790abdc-a796-482f-a40d-f8ee035ecfc2","ref_index":3,"cited_arxiv_id":"2410.24164","is_internal_anchor":true},{"doi":"","year":2025,"title":"$\\pi_{0.5}$: a Vision-Language-Action Model with Open-World Generalization","work_id":"d1ad7304-d09a-49bc-809e-846439f6aff9","ref_index":4,"cited_arxiv_id":"2504.16054","is_internal_anchor":true},{"doi":"","year":2025,"title":"SAM 3: Segment Anything with Concepts","work_id":"4a72a006-2592-4554-aad0-a9c41a9f952d","ref_index":5,"cited_arxiv_id":"2511.16719","is_internal_anchor":true}],"resolved_work":39,"snapshot_sha256":"22cd1f886b762cf19ce42cf5c43a1180b6db3330ac088ce97e4e4366502c125a","internal_anchors":18},"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"}