{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:P4I6LS2TSF4SSO2MTYQXHKOMQQ","short_pith_number":"pith:P4I6LS2T","schema_version":"1.0","canonical_sha256":"7f11e5cb539179293b4c9e2173a9cc840a7d1eae9317f9cbb68b8843037472a1","source":{"kind":"arxiv","id":"2408.06481","version":2},"attestation_state":"computed","paper":{"title":"UniT: Data Efficient Tactile Representation with Generalization to Unseen Objects","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Cael Fitch, Dongyi Wang, Philip Glen Crandall, Raghava Uppuluri, Wan Shou, Xinwei Zhang, Yu She, Zhengtong Xu","submitted_at":"2024-08-12T20:29:09Z","abstract_excerpt":"UniT is an approach to tactile representation learning, using VQGAN to learn a compact latent space and serve as the tactile representation. It uses tactile images obtained from a single simple object to train the representation with generalizability. This tactile representation can be zero-shot transferred to various downstream tasks, including perception tasks and manipulation policy learning. Our benchmarkings on in-hand 3D pose and 6D pose estimation tasks and a tactile classification task show that UniT outperforms existing visual and tactile representation learning methods. Additionally,"},"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":"2408.06481","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2024-08-12T20:29:09Z","cross_cats_sorted":[],"title_canon_sha256":"c99ff9db0749b4b900a8c16ae9b97da3c94978b9b918960bfa98b7a9524249a1","abstract_canon_sha256":"84c77759ea66ee84a375e4c0710de4a2a8477311211bdd0d529cbbe66bcf631c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:43:00.703217Z","signature_b64":"RBjwWwJY+Kt/sxqDPX5qExRBgEbPfXgyWzpgpQVoPSR3FSZ1vEErkHPxx3fUOO5PFy421HPvlR58bCWrwipkDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7f11e5cb539179293b4c9e2173a9cc840a7d1eae9317f9cbb68b8843037472a1","last_reissued_at":"2026-07-05T10:43:00.702516Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:43:00.702516Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"UniT: Data Efficient Tactile Representation with Generalization to Unseen Objects","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Cael Fitch, Dongyi Wang, Philip Glen Crandall, Raghava Uppuluri, Wan Shou, Xinwei Zhang, Yu She, Zhengtong Xu","submitted_at":"2024-08-12T20:29:09Z","abstract_excerpt":"UniT is an approach to tactile representation learning, using VQGAN to learn a compact latent space and serve as the tactile representation. It uses tactile images obtained from a single simple object to train the representation with generalizability. This tactile representation can be zero-shot transferred to various downstream tasks, including perception tasks and manipulation policy learning. Our benchmarkings on in-hand 3D pose and 6D pose estimation tasks and a tactile classification task show that UniT outperforms existing visual and tactile representation learning methods. Additionally,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.06481","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/2408.06481/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":"2408.06481","created_at":"2026-07-05T10:43:00.702586+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.06481v2","created_at":"2026-07-05T10:43:00.702586+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.06481","created_at":"2026-07-05T10:43:00.702586+00:00"},{"alias_kind":"pith_short_12","alias_value":"P4I6LS2TSF4S","created_at":"2026-07-05T10:43:00.702586+00:00"},{"alias_kind":"pith_short_16","alias_value":"P4I6LS2TSF4SSO2M","created_at":"2026-07-05T10:43:00.702586+00:00"},{"alias_kind":"pith_short_8","alias_value":"P4I6LS2T","created_at":"2026-07-05T10:43:00.702586+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26663","citing_title":"Tactile-WAM: Touch-Aware World Action Model with Tactile Asymmetric Attention","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2607.01684","citing_title":"Imagining the Sense of Touch: Touch-Informed Manipulation via Imagined Tactile Representations","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2606.31236","citing_title":"TactX: Learning Shared Tactile Representations Across Diverse Sensors","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2506.15953","citing_title":"ViTacFormer: Learning Cross-Modal Representation for Visuo-Tactile Dexterous Manipulation","ref_index":37,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/P4I6LS2TSF4SSO2MTYQXHKOMQQ","json":"https://pith.science/pith/P4I6LS2TSF4SSO2MTYQXHKOMQQ.json","graph_json":"https://pith.science/api/pith-number/P4I6LS2TSF4SSO2MTYQXHKOMQQ/graph.json","events_json":"https://pith.science/api/pith-number/P4I6LS2TSF4SSO2MTYQXHKOMQQ/events.json","paper":"https://pith.science/paper/P4I6LS2T"},"agent_actions":{"view_html":"https://pith.science/pith/P4I6LS2TSF4SSO2MTYQXHKOMQQ","download_json":"https://pith.science/pith/P4I6LS2TSF4SSO2MTYQXHKOMQQ.json","view_paper":"https://pith.science/paper/P4I6LS2T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.06481&json=true","fetch_graph":"https://pith.science/api/pith-number/P4I6LS2TSF4SSO2MTYQXHKOMQQ/graph.json","fetch_events":"https://pith.science/api/pith-number/P4I6LS2TSF4SSO2MTYQXHKOMQQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/P4I6LS2TSF4SSO2MTYQXHKOMQQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/P4I6LS2TSF4SSO2MTYQXHKOMQQ/action/storage_attestation","attest_author":"https://pith.science/pith/P4I6LS2TSF4SSO2MTYQXHKOMQQ/action/author_attestation","sign_citation":"https://pith.science/pith/P4I6LS2TSF4SSO2MTYQXHKOMQQ/action/citation_signature","submit_replication":"https://pith.science/pith/P4I6LS2TSF4SSO2MTYQXHKOMQQ/action/replication_record"}},"created_at":"2026-07-05T10:43:00.702586+00:00","updated_at":"2026-07-05T10:43:00.702586+00:00"}