{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:2QIT3AKDTQVI6XO2XDK5J2WNJV","short_pith_number":"pith:2QIT3AKD","schema_version":"1.0","canonical_sha256":"d4113d81439c2a8f5ddab8d5d4eacd4d5351d47c12071ed6b772d6ea3d8fab4e","source":{"kind":"arxiv","id":"2506.16211","version":1},"attestation_state":"computed","paper":{"title":"ControlVLA: Few-shot Object-centric Adaptation for Pre-trained Vision-Language-Action Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Jianan Wang, Puhao Li, Siyuan Huang, Song-Chun Zhu, Tengyu Liu, Wanlin Li, Yinghan Chen, Yingying Wu, Yuzhe Huang, Zhiyuan Zhang, Ziheng Xi","submitted_at":"2025-06-19T10:59:53Z","abstract_excerpt":"Learning real-world robotic manipulation is challenging, particularly when limited demonstrations are available. Existing methods for few-shot manipulation often rely on simulation-augmented data or pre-built modules like grasping and pose estimation, which struggle with sim-to-real gaps and lack extensibility. While large-scale imitation pre-training shows promise, adapting these general-purpose policies to specific tasks in data-scarce settings remains unexplored. To achieve this, we propose ControlVLA, a novel framework that bridges pre-trained VLA models with object-centric representations"},"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":"2506.16211","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2025-06-19T10:59:53Z","cross_cats_sorted":[],"title_canon_sha256":"131d93659668d80cbec60ecf99764315a6327175b234397560198f7002347f23","abstract_canon_sha256":"f413b68bc17ee64f9a4c5c05c7aeb447ac079065ee9cb0da84e01bb266194d8c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:24:32.863934Z","signature_b64":"TSmLZdSSTRZygvNQv2tg8aM7AteVhHsjHrcwGXCxfNJIl2Wqb0ZVCadujEEsYAxBVUkcxZEJ2SjzGaV3X2e7CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d4113d81439c2a8f5ddab8d5d4eacd4d5351d47c12071ed6b772d6ea3d8fab4e","last_reissued_at":"2026-07-05T11:24:32.863529Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:24:32.863529Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ControlVLA: Few-shot Object-centric Adaptation for Pre-trained Vision-Language-Action Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Jianan Wang, Puhao Li, Siyuan Huang, Song-Chun Zhu, Tengyu Liu, Wanlin Li, Yinghan Chen, Yingying Wu, Yuzhe Huang, Zhiyuan Zhang, Ziheng Xi","submitted_at":"2025-06-19T10:59:53Z","abstract_excerpt":"Learning real-world robotic manipulation is challenging, particularly when limited demonstrations are available. Existing methods for few-shot manipulation often rely on simulation-augmented data or pre-built modules like grasping and pose estimation, which struggle with sim-to-real gaps and lack extensibility. While large-scale imitation pre-training shows promise, adapting these general-purpose policies to specific tasks in data-scarce settings remains unexplored. To achieve this, we propose ControlVLA, a novel framework that bridges pre-trained VLA models with object-centric representations"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.16211","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2506.16211/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":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":"2506.16211","created_at":"2026-07-05T11:24:32.863581+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.16211v1","created_at":"2026-07-05T11:24:32.863581+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.16211","created_at":"2026-07-05T11:24:32.863581+00:00"},{"alias_kind":"pith_short_12","alias_value":"2QIT3AKDTQVI","created_at":"2026-07-05T11:24:32.863581+00:00"},{"alias_kind":"pith_short_16","alias_value":"2QIT3AKDTQVI6XO2","created_at":"2026-07-05T11:24:32.863581+00:00"},{"alias_kind":"pith_short_8","alias_value":"2QIT3AKD","created_at":"2026-07-05T11:24:32.863581+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":9,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.20867","citing_title":"FOCA: Future-Oriented Conditioning for Data-Efficient Vision-Language-Action Adaptation","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2606.13515","citing_title":"MaskWAM: Unifying Mask Prompting and Prediction for World-Action Models","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2606.14752","citing_title":"X-Tokenizer: A Multimodal Action Tokenizer for Vision-Language-Action Pretraining","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30484","citing_title":"ELAN4D: Embodiment-Centric 4D Supervision for Vision-Language-Action Models via Plug-and-Play Adaptation","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2511.06754","citing_title":"SlotVLA: Towards Modeling of Object-Relation Representations in Robotic Manipulation","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12236","citing_title":"TMRL: Diffusion Timestep-Modulated Pretraining Enables Exploration for Efficient Policy Finetuning","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2604.23121","citing_title":"Breaking Lock-In: Preserving Steerability under Low-Data VLA Post-Training","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05484","citing_title":"CoEnv: Driving Embodied Multi-Agent Collaboration via Compositional Environment","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17876","citing_title":"OFlow: Injecting Object-Aware Temporal Flow Matching for Robust Robotic Manipulation","ref_index":33,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2QIT3AKDTQVI6XO2XDK5J2WNJV","json":"https://pith.science/pith/2QIT3AKDTQVI6XO2XDK5J2WNJV.json","graph_json":"https://pith.science/api/pith-number/2QIT3AKDTQVI6XO2XDK5J2WNJV/graph.json","events_json":"https://pith.science/api/pith-number/2QIT3AKDTQVI6XO2XDK5J2WNJV/events.json","paper":"https://pith.science/paper/2QIT3AKD"},"agent_actions":{"view_html":"https://pith.science/pith/2QIT3AKDTQVI6XO2XDK5J2WNJV","download_json":"https://pith.science/pith/2QIT3AKDTQVI6XO2XDK5J2WNJV.json","view_paper":"https://pith.science/paper/2QIT3AKD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.16211&json=true","fetch_graph":"https://pith.science/api/pith-number/2QIT3AKDTQVI6XO2XDK5J2WNJV/graph.json","fetch_events":"https://pith.science/api/pith-number/2QIT3AKDTQVI6XO2XDK5J2WNJV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2QIT3AKDTQVI6XO2XDK5J2WNJV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2QIT3AKDTQVI6XO2XDK5J2WNJV/action/storage_attestation","attest_author":"https://pith.science/pith/2QIT3AKDTQVI6XO2XDK5J2WNJV/action/author_attestation","sign_citation":"https://pith.science/pith/2QIT3AKDTQVI6XO2XDK5J2WNJV/action/citation_signature","submit_replication":"https://pith.science/pith/2QIT3AKDTQVI6XO2XDK5J2WNJV/action/replication_record"}},"created_at":"2026-07-05T11:24:32.863581+00:00","updated_at":"2026-07-05T11:24:32.863581+00:00"}