{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:UEJ26R4ITDYNQTSX3LM7HKH2QH","short_pith_number":"pith:UEJ26R4I","schema_version":"1.0","canonical_sha256":"a113af478898f0d84e57dad9f3a8fa81c4ae4d379833f1e99d41fa6e67b5ebec","source":{"kind":"arxiv","id":"2507.17294","version":2},"attestation_state":"computed","paper":{"title":"VLA-Touch: Enhancing Vision-Language-Action Models with Dual-Level Tactile Feedback","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.RO","authors_text":"Ce Hao, Harold Soh, Jianxin Bi, Kevin Yuchen Ma, Mike Zheng Shou","submitted_at":"2025-07-23T07:54:10Z","abstract_excerpt":"Tactile feedback is generally recognized to be crucial for effective interaction with the physical world. However, state-of-the-art Vision-Language-Action (VLA) models lack the ability to interpret and use tactile signals, limiting their effectiveness in contact-rich tasks. Incorporating tactile feedback into these systems is challenging due to the absence of large multi-modal datasets. We present VLA-Touch, an approach that enhances generalist robot policies with tactile sensing \\emph{without fine-tuning} the base VLA. Our method introduces two key innovations: (1) a pipeline that leverages a"},"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":"2507.17294","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2025-07-23T07:54:10Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"4dae4c1a1cf735014f374bf43a9dcb98c7e7334274bd7c49dc7d04fbb0444b73","abstract_canon_sha256":"41cfed141ac8a750b4ca6f139ddbd4451e48842f316cc1482e99661976a9ef0b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:44:52.313487Z","signature_b64":"07MR0/R3QK/W19BGyI+gXcUf8FEn0Flfu/GR4HaRkiCrVpa+fnX1BOj650ETYLNALqpXT9h1gmaiBYQr0HN6BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a113af478898f0d84e57dad9f3a8fa81c4ae4d379833f1e99d41fa6e67b5ebec","last_reissued_at":"2026-07-05T11:44:52.313031Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:44:52.313031Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VLA-Touch: Enhancing Vision-Language-Action Models with Dual-Level Tactile Feedback","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.RO","authors_text":"Ce Hao, Harold Soh, Jianxin Bi, Kevin Yuchen Ma, Mike Zheng Shou","submitted_at":"2025-07-23T07:54:10Z","abstract_excerpt":"Tactile feedback is generally recognized to be crucial for effective interaction with the physical world. However, state-of-the-art Vision-Language-Action (VLA) models lack the ability to interpret and use tactile signals, limiting their effectiveness in contact-rich tasks. Incorporating tactile feedback into these systems is challenging due to the absence of large multi-modal datasets. We present VLA-Touch, an approach that enhances generalist robot policies with tactile sensing \\emph{without fine-tuning} the base VLA. Our method introduces two key innovations: (1) a pipeline that leverages a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.17294","kind":"arxiv","version":2},"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/2507.17294/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":"2507.17294","created_at":"2026-07-05T11:44:52.313087+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.17294v2","created_at":"2026-07-05T11:44:52.313087+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.17294","created_at":"2026-07-05T11:44:52.313087+00:00"},{"alias_kind":"pith_short_12","alias_value":"UEJ26R4ITDYN","created_at":"2026-07-05T11:44:52.313087+00:00"},{"alias_kind":"pith_short_16","alias_value":"UEJ26R4ITDYNQTSX","created_at":"2026-07-05T11:44:52.313087+00:00"},{"alias_kind":"pith_short_8","alias_value":"UEJ26R4I","created_at":"2026-07-05T11:44:52.313087+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":23,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.17598","citing_title":"MuseVLA: An Adaptive Multimodal Sensing Vision-Language-Action Model for Robotic Manipulation","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2606.13102","citing_title":"FTP-1: A Generalist Foundation Tactile Policy Across Tactile Sensors for Contact-Rich Manipulation","ref_index":74,"is_internal_anchor":false},{"citing_arxiv_id":"2606.11743","citing_title":"TacCoRL: Integrating Tactile Feedback into VLA via Simulation","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2606.11184","citing_title":"TacForeSight: Force-Guided Tactile World Model for Contact-Rich Manipulation","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2606.08737","citing_title":"Dream-Tac: A Unified Tactile World Action Model for Contact-Rich Robot Manipulation","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29173","citing_title":"TacGen: Touch Is a Necessary Dimension of Physical-World Representation -- Addressing Tactile Data Scarcity with Scalable Vision-to-Touch Alignment and Generation","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29941","citing_title":"Seeing Touch from Motion: A Unified Modality-Aware Visuo-Tactile Policy with Tactile Motion Correlation","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29173","citing_title":"TacGen: Touch Is a Necessary Dimension of Physical-World Representation -- Addressing Tactile Data Scarcity with Scalable Vision-to-Touch Alignment and Generation","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07308","citing_title":"AT-VLA: Adaptive Tactile Injection for Enhanced Feedback Reaction in Vision-Language-Action Models","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17336","citing_title":"Tactile-based Multimodal Fusion in Embodied Intelligence: A Survey of Vision, Language, and Contact-Driven Paradigms","ref_index":96,"is_internal_anchor":false},{"citing_arxiv_id":"2508.13073","citing_title":"Large VLM-based Vision-Language-Action Models for Robotic Manipulation: A Survey","ref_index":172,"is_internal_anchor":false},{"citing_arxiv_id":"2601.20239","citing_title":"TouchGuide: Inference-Time Steering of Visuomotor Policies via Touch Guidance","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2602.13833","citing_title":"Semantic-Contact Fields for Category-Level Generalizable Tactile Tool Manipulation","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04161","citing_title":"Adaptive Action Chunking at Inference-time for Vision-Language-Action Models","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02600","citing_title":"CoRAL: Contact-Rich Adaptive LLM-based Control for Robotic Manipulation","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03269","citing_title":"RLDX-1 Technical Report","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.23272","citing_title":"Modular Sensory Stream for Integrating Physical Feedback in Vision-Language-Action Models","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04649","citing_title":"From Reach to Insert: Tactile-Augmented Precision Assembly under Sub-Millimeter Tolerances","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2604.13015","citing_title":"Learning Versatile Humanoid Manipulation with Touch Dreaming","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07308","citing_title":"AT-VLA: Adaptive Tactile Injection for Enhanced Feedback Reaction in Vision-Language-Action Models","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04834","citing_title":"E-VLA: Event-Augmented Vision-Language-Action Model for Dark and Blurred Scenes","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02600","citing_title":"CoRAL: Contact-Rich Adaptive LLM-based Control for Robotic Manipulation","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03269","citing_title":"RLDX-1 Technical Report","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UEJ26R4ITDYNQTSX3LM7HKH2QH","json":"https://pith.science/pith/UEJ26R4ITDYNQTSX3LM7HKH2QH.json","graph_json":"https://pith.science/api/pith-number/UEJ26R4ITDYNQTSX3LM7HKH2QH/graph.json","events_json":"https://pith.science/api/pith-number/UEJ26R4ITDYNQTSX3LM7HKH2QH/events.json","paper":"https://pith.science/paper/UEJ26R4I"},"agent_actions":{"view_html":"https://pith.science/pith/UEJ26R4ITDYNQTSX3LM7HKH2QH","download_json":"https://pith.science/pith/UEJ26R4ITDYNQTSX3LM7HKH2QH.json","view_paper":"https://pith.science/paper/UEJ26R4I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.17294&json=true","fetch_graph":"https://pith.science/api/pith-number/UEJ26R4ITDYNQTSX3LM7HKH2QH/graph.json","fetch_events":"https://pith.science/api/pith-number/UEJ26R4ITDYNQTSX3LM7HKH2QH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UEJ26R4ITDYNQTSX3LM7HKH2QH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UEJ26R4ITDYNQTSX3LM7HKH2QH/action/storage_attestation","attest_author":"https://pith.science/pith/UEJ26R4ITDYNQTSX3LM7HKH2QH/action/author_attestation","sign_citation":"https://pith.science/pith/UEJ26R4ITDYNQTSX3LM7HKH2QH/action/citation_signature","submit_replication":"https://pith.science/pith/UEJ26R4ITDYNQTSX3LM7HKH2QH/action/replication_record"}},"created_at":"2026-07-05T11:44:52.313087+00:00","updated_at":"2026-07-05T11:44:52.313087+00:00"}