{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:OHCWK6DIAKVCY5JJEQK3OV7ZVZ","short_pith_number":"pith:OHCWK6DI","schema_version":"1.0","canonical_sha256":"71c565786802aa2c75292415b757f9ae7a21e440f231b2a4ccc99857b30bc588","source":{"kind":"arxiv","id":"2505.13441","version":3},"attestation_state":"computed","paper":{"title":"GraspMolmo: Generalizable Task-Oriented Grasping via Large-Scale Synthetic Data Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Abhay Deshpande, Arijit Ray, Jiafei Duan, Jordi Salvador, Kuo-Hao Zeng, Ranjay Krishna, Rose Hendrix, Winson Han, Yuke Zhu, Yuquan Deng","submitted_at":"2025-05-19T17:59:06Z","abstract_excerpt":"We present GrasMolmo, a generalizable open-vocabulary task-oriented grasping (TOG) model. GraspMolmo predicts semantically appropriate, stable grasps conditioned on a natural language instruction and a single RGB-D frame. For instance, given \"pour me some tea\", GraspMolmo selects a grasp on a teapot handle rather than its body. Unlike prior TOG methods, which are limited by small datasets, simplistic language, and uncluttered scenes, GraspMolmo learns from PRISM, a novel large-scale synthetic dataset of 379k samples featuring cluttered environments and diverse, realistic task descriptions. We "},"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":"2505.13441","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2025-05-19T17:59:06Z","cross_cats_sorted":[],"title_canon_sha256":"e45fb016ebddb68974206ef4ef41ce380344213957eedea539ed2ed6953abf0a","abstract_canon_sha256":"098843f2c81eb9f5484337a21109ae98461ff04f9228ecb41e5cc6ee719ba17f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:11:18.031842Z","signature_b64":"E1HnJ5U0HkeKEmQzsWZB/apVCyOTfud43kUGx9TqFxwbwJ26at5afwlTf29+OQF87bN9uqSb++tVvqNc6bUhBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"71c565786802aa2c75292415b757f9ae7a21e440f231b2a4ccc99857b30bc588","last_reissued_at":"2026-07-05T12:11:18.031348Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:11:18.031348Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GraspMolmo: Generalizable Task-Oriented Grasping via Large-Scale Synthetic Data Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Abhay Deshpande, Arijit Ray, Jiafei Duan, Jordi Salvador, Kuo-Hao Zeng, Ranjay Krishna, Rose Hendrix, Winson Han, Yuke Zhu, Yuquan Deng","submitted_at":"2025-05-19T17:59:06Z","abstract_excerpt":"We present GrasMolmo, a generalizable open-vocabulary task-oriented grasping (TOG) model. GraspMolmo predicts semantically appropriate, stable grasps conditioned on a natural language instruction and a single RGB-D frame. For instance, given \"pour me some tea\", GraspMolmo selects a grasp on a teapot handle rather than its body. Unlike prior TOG methods, which are limited by small datasets, simplistic language, and uncluttered scenes, GraspMolmo learns from PRISM, a novel large-scale synthetic dataset of 379k samples featuring cluttered environments and diverse, realistic task descriptions. We "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.13441","kind":"arxiv","version":3},"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/2505.13441/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":"2505.13441","created_at":"2026-07-05T12:11:18.031404+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.13441v3","created_at":"2026-07-05T12:11:18.031404+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.13441","created_at":"2026-07-05T12:11:18.031404+00:00"},{"alias_kind":"pith_short_12","alias_value":"OHCWK6DIAKVC","created_at":"2026-07-05T12:11:18.031404+00:00"},{"alias_kind":"pith_short_16","alias_value":"OHCWK6DIAKVCY5JJ","created_at":"2026-07-05T12:11:18.031404+00:00"},{"alias_kind":"pith_short_8","alias_value":"OHCWK6DI","created_at":"2026-07-05T12:11:18.031404+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.11324","citing_title":"Embodied-R1.5: Evolving Physical Intelligence via Embodied Foundation Models","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06155","citing_title":"AffordanceVLA: A Vision-Language-Action Model Empowering Action Generation through Affordance-Aware Understanding","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30161","citing_title":"Why Far Looks Up: Probing Spatial Representation in Vision-Language Models","ref_index":19,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OHCWK6DIAKVCY5JJEQK3OV7ZVZ","json":"https://pith.science/pith/OHCWK6DIAKVCY5JJEQK3OV7ZVZ.json","graph_json":"https://pith.science/api/pith-number/OHCWK6DIAKVCY5JJEQK3OV7ZVZ/graph.json","events_json":"https://pith.science/api/pith-number/OHCWK6DIAKVCY5JJEQK3OV7ZVZ/events.json","paper":"https://pith.science/paper/OHCWK6DI"},"agent_actions":{"view_html":"https://pith.science/pith/OHCWK6DIAKVCY5JJEQK3OV7ZVZ","download_json":"https://pith.science/pith/OHCWK6DIAKVCY5JJEQK3OV7ZVZ.json","view_paper":"https://pith.science/paper/OHCWK6DI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.13441&json=true","fetch_graph":"https://pith.science/api/pith-number/OHCWK6DIAKVCY5JJEQK3OV7ZVZ/graph.json","fetch_events":"https://pith.science/api/pith-number/OHCWK6DIAKVCY5JJEQK3OV7ZVZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OHCWK6DIAKVCY5JJEQK3OV7ZVZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OHCWK6DIAKVCY5JJEQK3OV7ZVZ/action/storage_attestation","attest_author":"https://pith.science/pith/OHCWK6DIAKVCY5JJEQK3OV7ZVZ/action/author_attestation","sign_citation":"https://pith.science/pith/OHCWK6DIAKVCY5JJEQK3OV7ZVZ/action/citation_signature","submit_replication":"https://pith.science/pith/OHCWK6DIAKVCY5JJEQK3OV7ZVZ/action/replication_record"}},"created_at":"2026-07-05T12:11:18.031404+00:00","updated_at":"2026-07-05T12:11:18.031404+00:00"}