{"paper":{"title":"HandelBot: Real-World Piano Playing via Fast Adaptation of Dexterous Robot Policies","license":"http://creativecommons.org/licenses/by/4.0/","headline":"HandelBot adapts a simulation policy in two stages to let a dexterous robot play piano accurately after 30 minutes of real data.","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Amber Xie, Dorsa Sadigh, Haozhi Qi","submitted_at":"2026-03-12T17:56:29Z","abstract_excerpt":"Mastering dexterous manipulation with multi-fingered hands has been a grand challenge in robotics for decades. Despite its potential, the difficulty of collecting high-quality data remains a primary bottleneck for high-precision tasks. While reinforcement learning and simulation-to-real-world transfer offer a promising alternative, the transferred policies often fail for tasks demanding millimeter-scale precision, such as bimanual piano playing. In this work, we introduce HandelBot, a framework that combines a simulation policy and rapid adaptation through a two-stage pipeline. Starting from a"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Through extensive hardware experiments across five recognized songs, we demonstrate that HandelBot can successfully perform precise bimanual piano playing. Our system outperforms direct simulation deployment by a factor of 1.8x and requires only 30 minutes of physical interaction data.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That a simulation-trained policy provides a sufficiently close starting point for the structured refinement stage to correct spatial misalignments to millimeter precision using only limited physical rollouts without introducing new instabilities in bimanual coordination.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"HandelBot achieves precise bimanual piano playing by refining a simulation policy through lateral finger adjustments and residual RL, outperforming direct sim deployment by 1.8x with only 30 minutes of physical data across five songs.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"HandelBot adapts a simulation policy in two stages to let a dexterous robot play piano accurately after 30 minutes of real data.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"21fd8c0c1f441cf68984ff2ad4d63af70cf43059fe47a8aee850d2f41c323515"},"source":{"id":"2603.12243","kind":"arxiv","version":4},"verdict":{"id":"3c5b6be6-2a2a-4f52-8aa1-a643c321fb02","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T11:45:02.148110Z","strongest_claim":"Through extensive hardware experiments across five recognized songs, we demonstrate that HandelBot can successfully perform precise bimanual piano playing. Our system outperforms direct simulation deployment by a factor of 1.8x and requires only 30 minutes of physical interaction data.","one_line_summary":"HandelBot achieves precise bimanual piano playing by refining a simulation policy through lateral finger adjustments and residual RL, outperforming direct sim deployment by 1.8x with only 30 minutes of physical data across five songs.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That a simulation-trained policy provides a sufficiently close starting point for the structured refinement stage to correct spatial misalignments to millimeter precision using only limited physical rollouts without introducing new instabilities in bimanual coordination.","pith_extraction_headline":"HandelBot adapts a simulation policy in two stages to let a dexterous robot play piano accurately after 30 minutes of real data."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2603.12243/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":2,"snapshot_sha256":"b4481a22bb3f504b786eae7b262dea5a80245b774a2be7314c6913984bd46814"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}