{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:MTKWJDZIKX6CLG3UPANYGDNCD3","short_pith_number":"pith:MTKWJDZI","schema_version":"1.0","canonical_sha256":"64d5648f2855fc259b74781b830da21ed4a344393e11bdf5beac004eef0cb8b0","source":{"kind":"arxiv","id":"2508.04199","version":1},"attestation_state":"computed","paper":{"title":"Reasoning Beyond Labels: Measuring LLM Sentiment in Low-Resource, Culturally Nuanced Contexts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Anja Thieme, Ignatius Ezeani, Jacki O'Neill, Javier Gonzalez, Keshet Ronen, Millicent Ochieng, Risa Ueno, Samuel Maina","submitted_at":"2025-08-06T08:27:55Z","abstract_excerpt":"Sentiment analysis in low-resource, culturally nuanced contexts challenges conventional NLP approaches that assume fixed labels and universal affective expressions. We present a diagnostic framework that treats sentiment as a context-dependent, culturally embedded construct, and evaluate how large language models (LLMs) reason about sentiment in informal, code-mixed WhatsApp messages from Nairobi youth health groups. Using a combination of human-annotated data, sentiment-flipped counterfactuals, and rubric-based explanation evaluation, we probe LLM interpretability, robustness, and alignment w"},"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":"2508.04199","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-06T08:27:55Z","cross_cats_sorted":[],"title_canon_sha256":"e02325235d5f65e1b1b4ea5f9c02953879152de24d3a34670a7079ce0c4f15fd","abstract_canon_sha256":"4e00367677ed739ec1347de48fcb022fadc72fbd8a04c520ddaca63f9cd6c270"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:49:22.667432Z","signature_b64":"HM2wDmcgr0LmNiPra7FHaCu6+BYVkJFEMc3DDhjUaEUPi4E0qj9H4Vmo4BJE0tcn3WDtZI0+CebcliRzN9cPDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"64d5648f2855fc259b74781b830da21ed4a344393e11bdf5beac004eef0cb8b0","last_reissued_at":"2026-07-05T11:49:22.666971Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:49:22.666971Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reasoning Beyond Labels: Measuring LLM Sentiment in Low-Resource, Culturally Nuanced Contexts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Anja Thieme, Ignatius Ezeani, Jacki O'Neill, Javier Gonzalez, Keshet Ronen, Millicent Ochieng, Risa Ueno, Samuel Maina","submitted_at":"2025-08-06T08:27:55Z","abstract_excerpt":"Sentiment analysis in low-resource, culturally nuanced contexts challenges conventional NLP approaches that assume fixed labels and universal affective expressions. We present a diagnostic framework that treats sentiment as a context-dependent, culturally embedded construct, and evaluate how large language models (LLMs) reason about sentiment in informal, code-mixed WhatsApp messages from Nairobi youth health groups. Using a combination of human-annotated data, sentiment-flipped counterfactuals, and rubric-based explanation evaluation, we probe LLM interpretability, robustness, and alignment w"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.04199","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/2508.04199/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":"2508.04199","created_at":"2026-07-05T11:49:22.667027+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.04199v1","created_at":"2026-07-05T11:49:22.667027+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.04199","created_at":"2026-07-05T11:49:22.667027+00:00"},{"alias_kind":"pith_short_12","alias_value":"MTKWJDZIKX6C","created_at":"2026-07-05T11:49:22.667027+00:00"},{"alias_kind":"pith_short_16","alias_value":"MTKWJDZIKX6CLG3U","created_at":"2026-07-05T11:49:22.667027+00:00"},{"alias_kind":"pith_short_8","alias_value":"MTKWJDZI","created_at":"2026-07-05T11:49:22.667027+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.20255","citing_title":"The Register Gap: A Meaning Intelligence Framework for Nigerian Public Discourse","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2602.17469","citing_title":"Cross-Lingual Sentiment Misalignment: Auditing Multilingual Language Models for Inversion Risk, Dialectal Representation, and Affective Stability","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MTKWJDZIKX6CLG3UPANYGDNCD3","json":"https://pith.science/pith/MTKWJDZIKX6CLG3UPANYGDNCD3.json","graph_json":"https://pith.science/api/pith-number/MTKWJDZIKX6CLG3UPANYGDNCD3/graph.json","events_json":"https://pith.science/api/pith-number/MTKWJDZIKX6CLG3UPANYGDNCD3/events.json","paper":"https://pith.science/paper/MTKWJDZI"},"agent_actions":{"view_html":"https://pith.science/pith/MTKWJDZIKX6CLG3UPANYGDNCD3","download_json":"https://pith.science/pith/MTKWJDZIKX6CLG3UPANYGDNCD3.json","view_paper":"https://pith.science/paper/MTKWJDZI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.04199&json=true","fetch_graph":"https://pith.science/api/pith-number/MTKWJDZIKX6CLG3UPANYGDNCD3/graph.json","fetch_events":"https://pith.science/api/pith-number/MTKWJDZIKX6CLG3UPANYGDNCD3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MTKWJDZIKX6CLG3UPANYGDNCD3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MTKWJDZIKX6CLG3UPANYGDNCD3/action/storage_attestation","attest_author":"https://pith.science/pith/MTKWJDZIKX6CLG3UPANYGDNCD3/action/author_attestation","sign_citation":"https://pith.science/pith/MTKWJDZIKX6CLG3UPANYGDNCD3/action/citation_signature","submit_replication":"https://pith.science/pith/MTKWJDZIKX6CLG3UPANYGDNCD3/action/replication_record"}},"created_at":"2026-07-05T11:49:22.667027+00:00","updated_at":"2026-07-05T11:49:22.667027+00:00"}