{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:Y4NHYBWTUNM4ZSY4P3JMJHC7E4","short_pith_number":"pith:Y4NHYBWT","schema_version":"1.0","canonical_sha256":"c71a7c06d3a359cccb1c7ed2c49c5f272ebfde8121a1aff8b284a868aace8433","source":{"kind":"arxiv","id":"2512.22385","version":2},"attestation_state":"computed","paper":{"title":"LLM-Guided Exemplar Selection for Few-Shot Wearable-Sensor Human Activity Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.CL","authors_text":"Elsen Ronando, Sozo Inoue","submitted_at":"2025-12-26T21:03:15Z","abstract_excerpt":"In this paper, we propose an LLM-Guided Exemplar Selection framework to address a key limitation in state-of-the-art Human Activity Recognition (HAR) methods: their reliance on large labeled datasets and purely geometric exemplar selection, which often fail to distinguish similar wearable sensor activities such as walking, walking upstairs, and walking downstairs. Our method incorporates semantic reasoning via an LLM-generated knowledge prior that captures feature importance, inter-class confusability, and exemplar budget multipliers, and uses it to guide exemplar scoring and selection. These "},"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":"2512.22385","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-12-26T21:03:15Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"e18d63e22958cbf69d823be21e7bb7c89faab3f4fa3a34e9d8df6e83cd785f0f","abstract_canon_sha256":"9b3728bb2aa50235e5759b3916c8f0ffacec8a9a7de0e71726b8cecc7daf173f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-21T01:19:51.598325Z","signature_b64":"+TXuruVcoCKtXaGNumwSvnJ3k0cfh9CAA7j4Nt/LuqMBY0u3U374bYYjbVz9Am8kA2EuY3quqpL+bzcyKZmVBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c71a7c06d3a359cccb1c7ed2c49c5f272ebfde8121a1aff8b284a868aace8433","last_reissued_at":"2026-07-21T01:19:51.597386Z","signature_status":"signed_v1","first_computed_at":"2026-07-21T01:19:51.597386Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LLM-Guided Exemplar Selection for Few-Shot Wearable-Sensor Human Activity Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.CL","authors_text":"Elsen Ronando, Sozo Inoue","submitted_at":"2025-12-26T21:03:15Z","abstract_excerpt":"In this paper, we propose an LLM-Guided Exemplar Selection framework to address a key limitation in state-of-the-art Human Activity Recognition (HAR) methods: their reliance on large labeled datasets and purely geometric exemplar selection, which often fail to distinguish similar wearable sensor activities such as walking, walking upstairs, and walking downstairs. Our method incorporates semantic reasoning via an LLM-generated knowledge prior that captures feature importance, inter-class confusability, and exemplar budget multipliers, and uses it to guide exemplar scoring and selection. These "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2512.22385","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/2512.22385/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":"2512.22385","created_at":"2026-07-21T01:19:51.597828+00:00"},{"alias_kind":"arxiv_version","alias_value":"2512.22385v2","created_at":"2026-07-21T01:19:51.597828+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.22385","created_at":"2026-07-21T01:19:51.597828+00:00"},{"alias_kind":"pith_short_12","alias_value":"Y4NHYBWTUNM4","created_at":"2026-07-21T01:19:51.597828+00:00"},{"alias_kind":"pith_short_16","alias_value":"Y4NHYBWTUNM4ZSY4","created_at":"2026-07-21T01:19:51.597828+00:00"},{"alias_kind":"pith_short_8","alias_value":"Y4NHYBWT","created_at":"2026-07-21T01:19:51.597828+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.02841","citing_title":"TRACE: Temporal Reasoning over Context and Evidence for Activity Recognition in Smart Homes","ref_index":73,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Y4NHYBWTUNM4ZSY4P3JMJHC7E4","json":"https://pith.science/pith/Y4NHYBWTUNM4ZSY4P3JMJHC7E4.json","graph_json":"https://pith.science/api/pith-number/Y4NHYBWTUNM4ZSY4P3JMJHC7E4/graph.json","events_json":"https://pith.science/api/pith-number/Y4NHYBWTUNM4ZSY4P3JMJHC7E4/events.json","paper":"https://pith.science/paper/Y4NHYBWT"},"agent_actions":{"view_html":"https://pith.science/pith/Y4NHYBWTUNM4ZSY4P3JMJHC7E4","download_json":"https://pith.science/pith/Y4NHYBWTUNM4ZSY4P3JMJHC7E4.json","view_paper":"https://pith.science/paper/Y4NHYBWT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2512.22385&json=true","fetch_graph":"https://pith.science/api/pith-number/Y4NHYBWTUNM4ZSY4P3JMJHC7E4/graph.json","fetch_events":"https://pith.science/api/pith-number/Y4NHYBWTUNM4ZSY4P3JMJHC7E4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Y4NHYBWTUNM4ZSY4P3JMJHC7E4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Y4NHYBWTUNM4ZSY4P3JMJHC7E4/action/storage_attestation","attest_author":"https://pith.science/pith/Y4NHYBWTUNM4ZSY4P3JMJHC7E4/action/author_attestation","sign_citation":"https://pith.science/pith/Y4NHYBWTUNM4ZSY4P3JMJHC7E4/action/citation_signature","submit_replication":"https://pith.science/pith/Y4NHYBWTUNM4ZSY4P3JMJHC7E4/action/replication_record"}},"created_at":"2026-07-21T01:19:51.597828+00:00","updated_at":"2026-07-21T01:19:51.597828+00:00"}