{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:HL3ZGID7PX4IB3AG6AW3FCRYPI","short_pith_number":"pith:HL3ZGID7","schema_version":"1.0","canonical_sha256":"3af793207f7df880ec06f02db28a387a3fec15465aea4a0e3c5f391545468730","source":{"kind":"arxiv","id":"2608.06898","version":1},"attestation_state":"computed","paper":{"title":"How Should I Pick a Foundation Model for My Robot? In Favor of a Community Evaluation Framework for Social Robots","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.HC"],"primary_cat":"cs.RO","authors_text":"Alva Markelius, Eric Nichols, Hatice Gunes","submitted_at":"2026-08-07T07:33:25Z","abstract_excerpt":"Researchers who seek to build social robot applications on foundation models are faced with a difficult question: how should we pick a model? Public leaderboards offer little guidance: the demands of real-time, embodied social interaction lie largely outside their focus. And direct evaluation is impractical at scale: each embodied study requires scarce participant, robot, and experimenter time. In this paper, we identify five evaluation dimensions for foundation models in social robots: (i) conversational competence, (ii) user safety, (iii) embodied character, (iv) target scene effectiveness, "},"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":"2608.06898","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-08-07T07:33:25Z","cross_cats_sorted":["cs.CL","cs.HC"],"title_canon_sha256":"c6f02321f1b5c89fa60b21d19a63c14e1b8e9c2d78b61be0f695d2131387bc70","abstract_canon_sha256":"d877379441154f30c6487e04f4ab310643507401b4877dbce0826b4d3238d173"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-10T01:12:27.038845Z","signature_b64":"AsEnD5q83eVqLcYAr6e3sf5CPK3arY18W48V/SivJN/l8DaqXlUnoHEt6/G7YX1BPUseNyVKWku3M115b7v3Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3af793207f7df880ec06f02db28a387a3fec15465aea4a0e3c5f391545468730","last_reissued_at":"2026-08-10T01:12:27.036423Z","signature_status":"signed_v1","first_computed_at":"2026-08-10T01:12:27.036423Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"How Should I Pick a Foundation Model for My Robot? In Favor of a Community Evaluation Framework for Social Robots","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.HC"],"primary_cat":"cs.RO","authors_text":"Alva Markelius, Eric Nichols, Hatice Gunes","submitted_at":"2026-08-07T07:33:25Z","abstract_excerpt":"Researchers who seek to build social robot applications on foundation models are faced with a difficult question: how should we pick a model? Public leaderboards offer little guidance: the demands of real-time, embodied social interaction lie largely outside their focus. And direct evaluation is impractical at scale: each embodied study requires scarce participant, robot, and experimenter time. In this paper, we identify five evaluation dimensions for foundation models in social robots: (i) conversational competence, (ii) user safety, (iii) embodied character, (iv) target scene effectiveness, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.06898","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/2608.06898/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":"2608.06898","created_at":"2026-08-10T01:12:27.037377+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.06898v1","created_at":"2026-08-10T01:12:27.037377+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.06898","created_at":"2026-08-10T01:12:27.037377+00:00"},{"alias_kind":"pith_short_12","alias_value":"HL3ZGID7PX4I","created_at":"2026-08-10T01:12:27.037377+00:00"},{"alias_kind":"pith_short_16","alias_value":"HL3ZGID7PX4IB3AG","created_at":"2026-08-10T01:12:27.037377+00:00"},{"alias_kind":"pith_short_8","alias_value":"HL3ZGID7","created_at":"2026-08-10T01:12:27.037377+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/HL3ZGID7PX4IB3AG6AW3FCRYPI","json":"https://pith.science/pith/HL3ZGID7PX4IB3AG6AW3FCRYPI.json","graph_json":"https://pith.science/api/pith-number/HL3ZGID7PX4IB3AG6AW3FCRYPI/graph.json","events_json":"https://pith.science/api/pith-number/HL3ZGID7PX4IB3AG6AW3FCRYPI/events.json","paper":"https://pith.science/paper/HL3ZGID7"},"agent_actions":{"view_html":"https://pith.science/pith/HL3ZGID7PX4IB3AG6AW3FCRYPI","download_json":"https://pith.science/pith/HL3ZGID7PX4IB3AG6AW3FCRYPI.json","view_paper":"https://pith.science/paper/HL3ZGID7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.06898&json=true","fetch_graph":"https://pith.science/api/pith-number/HL3ZGID7PX4IB3AG6AW3FCRYPI/graph.json","fetch_events":"https://pith.science/api/pith-number/HL3ZGID7PX4IB3AG6AW3FCRYPI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HL3ZGID7PX4IB3AG6AW3FCRYPI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HL3ZGID7PX4IB3AG6AW3FCRYPI/action/storage_attestation","attest_author":"https://pith.science/pith/HL3ZGID7PX4IB3AG6AW3FCRYPI/action/author_attestation","sign_citation":"https://pith.science/pith/HL3ZGID7PX4IB3AG6AW3FCRYPI/action/citation_signature","submit_replication":"https://pith.science/pith/HL3ZGID7PX4IB3AG6AW3FCRYPI/action/replication_record"}},"created_at":"2026-08-10T01:12:27.037377+00:00","updated_at":"2026-08-10T01:12:27.037377+00:00"}