{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:XBQQK7XJ7UBAI3P36LGAGH2FQR","short_pith_number":"pith:XBQQK7XJ","schema_version":"1.0","canonical_sha256":"b861057ee9fd02046dfbf2cc031f458458a675035cf2b8512fce4f2ba82b0a7c","source":{"kind":"arxiv","id":"2509.05031","version":1},"attestation_state":"computed","paper":{"title":"Pointing-Guided Target Estimation via Transformer-Based Attention","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.RO","authors_text":"Hassan Ali, Luca M\\\"uller, Luk\\'a\\v{s} Gajdo\\v{s}ech, Philipp Allgeuer, Stefan Wermter","submitted_at":"2025-09-05T11:42:03Z","abstract_excerpt":"Deictic gestures, like pointing, are a fundamental form of non-verbal communication, enabling humans to direct attention to specific objects or locations. This capability is essential in Human-Robot Interaction (HRI), where robots should be able to predict human intent and anticipate appropriate responses. In this work, we propose the Multi-Modality Inter-TransFormer (MM-ITF), a modular architecture to predict objects in a controlled tabletop scenario with the NICOL robot, where humans indicate targets through natural pointing gestures. Leveraging inter-modality attention, MM-ITF maps 2D point"},"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":"2509.05031","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2025-09-05T11:42:03Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"41d4a5af33bb69fd78b6bd85f0fb18ee44c584430e9485a5979b56fd6aca53a7","abstract_canon_sha256":"0293a51437a170990ad98bdf777ac73bd7127f54b633d5940bced1e364b0bd45"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:05:34.784281Z","signature_b64":"4TaGRmnb/1PnD921oKPJgWG84eu1+uszwxuV324pOBdHaoXl8gFe698ATn+jdPXVu3T25r6iaGqkQVTnUqKPCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b861057ee9fd02046dfbf2cc031f458458a675035cf2b8512fce4f2ba82b0a7c","last_reissued_at":"2026-07-05T12:05:34.783808Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:05:34.783808Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Pointing-Guided Target Estimation via Transformer-Based Attention","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.RO","authors_text":"Hassan Ali, Luca M\\\"uller, Luk\\'a\\v{s} Gajdo\\v{s}ech, Philipp Allgeuer, Stefan Wermter","submitted_at":"2025-09-05T11:42:03Z","abstract_excerpt":"Deictic gestures, like pointing, are a fundamental form of non-verbal communication, enabling humans to direct attention to specific objects or locations. This capability is essential in Human-Robot Interaction (HRI), where robots should be able to predict human intent and anticipate appropriate responses. In this work, we propose the Multi-Modality Inter-TransFormer (MM-ITF), a modular architecture to predict objects in a controlled tabletop scenario with the NICOL robot, where humans indicate targets through natural pointing gestures. Leveraging inter-modality attention, MM-ITF maps 2D point"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.05031","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/2509.05031/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":"2509.05031","created_at":"2026-07-05T12:05:34.783860+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.05031v1","created_at":"2026-07-05T12:05:34.783860+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.05031","created_at":"2026-07-05T12:05:34.783860+00:00"},{"alias_kind":"pith_short_12","alias_value":"XBQQK7XJ7UBA","created_at":"2026-07-05T12:05:34.783860+00:00"},{"alias_kind":"pith_short_16","alias_value":"XBQQK7XJ7UBAI3P3","created_at":"2026-07-05T12:05:34.783860+00:00"},{"alias_kind":"pith_short_8","alias_value":"XBQQK7XJ","created_at":"2026-07-05T12:05:34.783860+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XBQQK7XJ7UBAI3P36LGAGH2FQR","json":"https://pith.science/pith/XBQQK7XJ7UBAI3P36LGAGH2FQR.json","graph_json":"https://pith.science/api/pith-number/XBQQK7XJ7UBAI3P36LGAGH2FQR/graph.json","events_json":"https://pith.science/api/pith-number/XBQQK7XJ7UBAI3P36LGAGH2FQR/events.json","paper":"https://pith.science/paper/XBQQK7XJ"},"agent_actions":{"view_html":"https://pith.science/pith/XBQQK7XJ7UBAI3P36LGAGH2FQR","download_json":"https://pith.science/pith/XBQQK7XJ7UBAI3P36LGAGH2FQR.json","view_paper":"https://pith.science/paper/XBQQK7XJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.05031&json=true","fetch_graph":"https://pith.science/api/pith-number/XBQQK7XJ7UBAI3P36LGAGH2FQR/graph.json","fetch_events":"https://pith.science/api/pith-number/XBQQK7XJ7UBAI3P36LGAGH2FQR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XBQQK7XJ7UBAI3P36LGAGH2FQR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XBQQK7XJ7UBAI3P36LGAGH2FQR/action/storage_attestation","attest_author":"https://pith.science/pith/XBQQK7XJ7UBAI3P36LGAGH2FQR/action/author_attestation","sign_citation":"https://pith.science/pith/XBQQK7XJ7UBAI3P36LGAGH2FQR/action/citation_signature","submit_replication":"https://pith.science/pith/XBQQK7XJ7UBAI3P36LGAGH2FQR/action/replication_record"}},"created_at":"2026-07-05T12:05:34.783860+00:00","updated_at":"2026-07-05T12:05:34.783860+00:00"}