{"paper":{"title":"When Absolute State Fails: Evaluating Proprioceptive Encodings for Robust Manipulation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A simple episode-wise relative frame for proprioceptive encoding delivers better performance and robustness than absolute state representations in real robotic manipulation.","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Afshin Zeinaddini Meymand, Genki Sano, Maxime Alvarez, Pablo Ferreiro, Paul Crook, Ryo Watanabe, Suvin Kurian","submitted_at":"2026-05-13T06:41:56Z","abstract_excerpt":"As end-to-end robotic policies are progressively deployed in the real world to solve real tasks, they face a gap between the training and inference conditions. Scaling the amount and diversity of the training data has shown some success in improving zero-shot generalization, yet robots still fail when faced with new, unseen test conditions. For instance, while robots with fixed frames of reference are common, those with moving frames pose a greater challenge for deployment. To address this specific instance of the issue, we present a study of strategies for encoding the robot's proprioceptive "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"a simple episode-wise relative frame provides the best trade-off between task performance and robustness, outperforming the baselines in extensive real-robot experiments conducted in a realistic test environment.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the chosen test environment and task variations sufficiently represent the broader range of frame changes encountered in real-world deployment.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Episode-wise relative proprioceptive encoding outperforms absolute state baselines for robust robotic manipulation under varying reference frames.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A simple episode-wise relative frame for proprioceptive encoding delivers better performance and robustness than absolute state representations in real robotic manipulation.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"30081b5fab6c25252d3e707fec1c1d46e39b0d12740a7c9c1cb981ff2606d2ba"},"source":{"id":"2605.13067","kind":"arxiv","version":1},"verdict":{"id":"41c2f189-4eb6-4702-8a9c-234e2bb696aa","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T19:03:32.513350Z","strongest_claim":"a simple episode-wise relative frame provides the best trade-off between task performance and robustness, outperforming the baselines in extensive real-robot experiments conducted in a realistic test environment.","one_line_summary":"Episode-wise relative proprioceptive encoding outperforms absolute state baselines for robust robotic manipulation under varying reference frames.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the chosen test environment and task variations sufficiently represent the broader range of frame changes encountered in real-world deployment.","pith_extraction_headline":"A simple episode-wise relative frame for proprioceptive encoding delivers better performance and robustness than absolute state representations in real robotic manipulation."},"references":{"count":19,"sample":[{"doi":"","year":2025,"title":"Airoa moma dataset: A large-scale hierarchical dataset for mobile manipulation,","work_id":"8284726c-46f8-49e9-a022-a373a39b6ffe","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Tidybot++: An open-source holonomic mobile manipulator for robot learning,","work_id":"e9522fa6-efbc-4de9-a62c-592a98bc5d09","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"An autonomous mobile robot navigation architecture for dynamic intralogistics,","work_id":"ddface24-8128-4c7b-b112-31a4e756e254","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1109/icra46639","year":2022,"title":"In: 2022 International Conference on Robotics and Automation (ICRA), pp","work_id":"0b0f4342-f6a2-45b1-afb7-916a866f4f25","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2026,"title":"Telexistence, “Ghost,” https://tx-inc.com/en/technology/, online; ac- cessed 13-Apr-2026","work_id":"e7164a57-966b-4452-8ddb-3d7e914cb46a","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":19,"snapshot_sha256":"fb9a0a4363b3657b7a4de55531c7bc1ae125553c694da84628927be30244563c","internal_anchors":1},"formal_canon":{"evidence_count":2,"snapshot_sha256":"123353b1f418fbec9f96ca5d0d48a189fe56b19789871ad8952686a81c5b6675"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}