{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:GHFBJ72HW77P35ZLCKFANMSKPO","short_pith_number":"pith:GHFBJ72H","schema_version":"1.0","canonical_sha256":"31ca14ff47b7fefdf72b128a06b24a7b9e0fd06dd47e3f6b2b371f55c8792588","source":{"kind":"arxiv","id":"2606.17160","version":1},"attestation_state":"computed","paper":{"title":"Transductive Zero-Shot Audio Classification with Audio-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.SD","authors_text":"Jingwen Zhou, Mingzhe Wang","submitted_at":"2026-06-15T18:01:08Z","abstract_excerpt":"Contrastive language-audio pretraining (CLAP) enables zero-shot audio classification, but standard inference classifies each clip in isolation and ignores the structure of the unlabeled test set. We present the first systematic study of TransCLIP-style transductive inference for CLAP: a text-anchored spherical Gaussian-mixture EM that refines zero-shot posteriors using the audio-embedding statistics of the test batch, with no labels, no gradients, and negligible compute (about 15 ms on one CPU core for 2,000 clips). Across ESC-50, UrbanSound8K, and VocalSound, this consistently improves top-1 "},"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":"2606.17160","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2026-06-15T18:01:08Z","cross_cats_sorted":[],"title_canon_sha256":"1e264cad5089da870415e3ac2e642d0e8d01ac86decc1e3bf9a5792561cec7c7","abstract_canon_sha256":"b63b5a31c147e5a2b529e55bb375213189d0a18d46593d218e730eb572dc6a51"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:10:04.707948Z","signature_b64":"DzCg5iHMiQH3Kx3DQyoDfp8uugZnmF9xQT6zPV4GRlTdgQz2ErvOTYNlCUFYRVltwsuN+KlvY1mR5+W37DrKDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"31ca14ff47b7fefdf72b128a06b24a7b9e0fd06dd47e3f6b2b371f55c8792588","last_reissued_at":"2026-06-19T16:10:04.707611Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:10:04.707611Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Transductive Zero-Shot Audio Classification with Audio-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.SD","authors_text":"Jingwen Zhou, Mingzhe Wang","submitted_at":"2026-06-15T18:01:08Z","abstract_excerpt":"Contrastive language-audio pretraining (CLAP) enables zero-shot audio classification, but standard inference classifies each clip in isolation and ignores the structure of the unlabeled test set. We present the first systematic study of TransCLIP-style transductive inference for CLAP: a text-anchored spherical Gaussian-mixture EM that refines zero-shot posteriors using the audio-embedding statistics of the test batch, with no labels, no gradients, and negligible compute (about 15 ms on one CPU core for 2,000 clips). Across ESC-50, UrbanSound8K, and VocalSound, this consistently improves top-1 "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.17160","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/2606.17160/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":"2606.17160","created_at":"2026-06-19T16:10:04.707671+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.17160v1","created_at":"2026-06-19T16:10:04.707671+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.17160","created_at":"2026-06-19T16:10:04.707671+00:00"},{"alias_kind":"pith_short_12","alias_value":"GHFBJ72HW77P","created_at":"2026-06-19T16:10:04.707671+00:00"},{"alias_kind":"pith_short_16","alias_value":"GHFBJ72HW77P35ZL","created_at":"2026-06-19T16:10:04.707671+00:00"},{"alias_kind":"pith_short_8","alias_value":"GHFBJ72H","created_at":"2026-06-19T16:10:04.707671+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/GHFBJ72HW77P35ZLCKFANMSKPO","json":"https://pith.science/pith/GHFBJ72HW77P35ZLCKFANMSKPO.json","graph_json":"https://pith.science/api/pith-number/GHFBJ72HW77P35ZLCKFANMSKPO/graph.json","events_json":"https://pith.science/api/pith-number/GHFBJ72HW77P35ZLCKFANMSKPO/events.json","paper":"https://pith.science/paper/GHFBJ72H"},"agent_actions":{"view_html":"https://pith.science/pith/GHFBJ72HW77P35ZLCKFANMSKPO","download_json":"https://pith.science/pith/GHFBJ72HW77P35ZLCKFANMSKPO.json","view_paper":"https://pith.science/paper/GHFBJ72H","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.17160&json=true","fetch_graph":"https://pith.science/api/pith-number/GHFBJ72HW77P35ZLCKFANMSKPO/graph.json","fetch_events":"https://pith.science/api/pith-number/GHFBJ72HW77P35ZLCKFANMSKPO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GHFBJ72HW77P35ZLCKFANMSKPO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GHFBJ72HW77P35ZLCKFANMSKPO/action/storage_attestation","attest_author":"https://pith.science/pith/GHFBJ72HW77P35ZLCKFANMSKPO/action/author_attestation","sign_citation":"https://pith.science/pith/GHFBJ72HW77P35ZLCKFANMSKPO/action/citation_signature","submit_replication":"https://pith.science/pith/GHFBJ72HW77P35ZLCKFANMSKPO/action/replication_record"}},"created_at":"2026-06-19T16:10:04.707671+00:00","updated_at":"2026-06-19T16:10:04.707671+00:00"}