{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YHWKXVB4E5X5N2VSTCWICM7OSJ","short_pith_number":"pith:YHWKXVB4","schema_version":"1.0","canonical_sha256":"c1ecabd43c276fd6eab298ac8133ee924893f11b30c82b044da954c250a8ff64","source":{"kind":"arxiv","id":"2505.14449","version":3},"attestation_state":"computed","paper":{"title":"Mitigating Subgroup Disparities in Multi-Label Speech Emotion Recognition: A Pseudo-Labeling and Unsupervised Learning Approach","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.CL","cs.SD"],"primary_cat":"eess.AS","authors_text":"Huang-Cheng Chou, Hung-yi Lee, Yi-Cheng Lin","submitted_at":"2025-05-20T14:50:44Z","abstract_excerpt":"While subgroup disparities and performance bias are increasingly studied in computational research, fairness in categorical Speech Emotion Recognition (SER) remains underexplored. Existing methods often rely on explicit demographic labels, which are difficult to obtain due to privacy concerns. To address this limitation, we introduce an Implicit Demography Inference (IDI) module that leverages pseudo-labeling from a pre-trained model and unsupervised learning using k-means clustering to mitigate bias in SER. Our experiments show that pseudo-labeling IDI reduces subgroup disparities, improving "},"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":"2505.14449","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"eess.AS","submitted_at":"2025-05-20T14:50:44Z","cross_cats_sorted":["cs.CL","cs.SD"],"title_canon_sha256":"76b59f8dcf66b05082694b6020eb3dcdec0cba6ad4956236b68b7cb8ff27e22f","abstract_canon_sha256":"2c942e27aab359d36678cca7a4a84a3c4ad0ce290ce810de8115b6c40677f973"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:32.350658Z","signature_b64":"5c8mkWtLqbHKa1KUOXPhvEmTkkBBGR0ybQpQRsYqcG7xs7qtsX0bhYUBVqLblnUGMnoKLKPEUF5pnj5BsApGBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c1ecabd43c276fd6eab298ac8133ee924893f11b30c82b044da954c250a8ff64","last_reissued_at":"2026-07-05T11:12:32.350087Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:32.350087Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Mitigating Subgroup Disparities in Multi-Label Speech Emotion Recognition: A Pseudo-Labeling and Unsupervised Learning Approach","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.CL","cs.SD"],"primary_cat":"eess.AS","authors_text":"Huang-Cheng Chou, Hung-yi Lee, Yi-Cheng Lin","submitted_at":"2025-05-20T14:50:44Z","abstract_excerpt":"While subgroup disparities and performance bias are increasingly studied in computational research, fairness in categorical Speech Emotion Recognition (SER) remains underexplored. Existing methods often rely on explicit demographic labels, which are difficult to obtain due to privacy concerns. To address this limitation, we introduce an Implicit Demography Inference (IDI) module that leverages pseudo-labeling from a pre-trained model and unsupervised learning using k-means clustering to mitigate bias in SER. Our experiments show that pseudo-labeling IDI reduces subgroup disparities, improving "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.14449","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/2505.14449/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":"2505.14449","created_at":"2026-07-05T11:12:32.350157+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.14449v3","created_at":"2026-07-05T11:12:32.350157+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.14449","created_at":"2026-07-05T11:12:32.350157+00:00"},{"alias_kind":"pith_short_12","alias_value":"YHWKXVB4E5X5","created_at":"2026-07-05T11:12:32.350157+00:00"},{"alias_kind":"pith_short_16","alias_value":"YHWKXVB4E5X5N2VS","created_at":"2026-07-05T11:12:32.350157+00:00"},{"alias_kind":"pith_short_8","alias_value":"YHWKXVB4","created_at":"2026-07-05T11:12:32.350157+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.11219","citing_title":"Afrispeech Semantics: Evaluating Audio Semantic Reasoning in Spoken Language Models Across Domains and Accents","ref_index":237,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11098","citing_title":"AffectCodec: Emotion-Preserving Neural Speech Codec for Expressive Speech Modeling","ref_index":89,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YHWKXVB4E5X5N2VSTCWICM7OSJ","json":"https://pith.science/pith/YHWKXVB4E5X5N2VSTCWICM7OSJ.json","graph_json":"https://pith.science/api/pith-number/YHWKXVB4E5X5N2VSTCWICM7OSJ/graph.json","events_json":"https://pith.science/api/pith-number/YHWKXVB4E5X5N2VSTCWICM7OSJ/events.json","paper":"https://pith.science/paper/YHWKXVB4"},"agent_actions":{"view_html":"https://pith.science/pith/YHWKXVB4E5X5N2VSTCWICM7OSJ","download_json":"https://pith.science/pith/YHWKXVB4E5X5N2VSTCWICM7OSJ.json","view_paper":"https://pith.science/paper/YHWKXVB4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.14449&json=true","fetch_graph":"https://pith.science/api/pith-number/YHWKXVB4E5X5N2VSTCWICM7OSJ/graph.json","fetch_events":"https://pith.science/api/pith-number/YHWKXVB4E5X5N2VSTCWICM7OSJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YHWKXVB4E5X5N2VSTCWICM7OSJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YHWKXVB4E5X5N2VSTCWICM7OSJ/action/storage_attestation","attest_author":"https://pith.science/pith/YHWKXVB4E5X5N2VSTCWICM7OSJ/action/author_attestation","sign_citation":"https://pith.science/pith/YHWKXVB4E5X5N2VSTCWICM7OSJ/action/citation_signature","submit_replication":"https://pith.science/pith/YHWKXVB4E5X5N2VSTCWICM7OSJ/action/replication_record"}},"created_at":"2026-07-05T11:12:32.350157+00:00","updated_at":"2026-07-05T11:12:32.350157+00:00"}