{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ETZAOYRG4KQ3R3VFPO6JBRNXAE","short_pith_number":"pith:ETZAOYRG","schema_version":"1.0","canonical_sha256":"24f2076226e2a1b8eea57bbc90c5b7011722e4c1f8ca587baea068a9e6471c8b","source":{"kind":"arxiv","id":"2605.13672","version":1},"attestation_state":"computed","paper":{"title":"SpurAudio: A Benchmark for Studying Shortcut Learning in Few-Shot Audio Classification","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A new benchmark shows few-shot audio classifiers suffer sharp drops when background correlations are broken, even in large pretrained models.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Giries Abu Ayoub, Loay Mualem, Morad Tukan","submitted_at":"2026-05-13T15:32:57Z","abstract_excerpt":"Few-shot classification (FSC) is widely used for learning from limited labeled data, yet most evaluations implicitly assume that target concepts are independent of contextual cues. In real-world settings, however, examples often appear within rich contexts, allowing models to exploit spurious correlations between foreground content and background signals. While such effects have been studied in few-shot image classification, their role in few-shot audio classification remains largely unexplored, and existing audio benchmarks offer limited control over contextual structure. We introduce SpurAud"},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2605.13672","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-13T15:32:57Z","cross_cats_sorted":[],"title_canon_sha256":"831c85f859235355edf55da23da924a86619c6a93fb02c04343a993b1b33dc8a","abstract_canon_sha256":"62c1ff2a536c654ae17342b7d78ac5388014e230cf35d67c6710aec7eb2d0ec2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T02:44:17.159147Z","signature_b64":"WDAK5XSh0GPViOsaBsTYIODGqRzkrhicwxIPwsHs8NeJ5QUJrCG6l3BwbqZPOaEmj7cUMZvywxgbv7FHE5ZsCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"24f2076226e2a1b8eea57bbc90c5b7011722e4c1f8ca587baea068a9e6471c8b","last_reissued_at":"2026-05-18T02:44:17.158648Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T02:44:17.158648Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SpurAudio: A Benchmark for Studying Shortcut Learning in Few-Shot Audio Classification","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A new benchmark shows few-shot audio classifiers suffer sharp drops when background correlations are broken, even in large pretrained models.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Giries Abu Ayoub, Loay Mualem, Morad Tukan","submitted_at":"2026-05-13T15:32:57Z","abstract_excerpt":"Few-shot classification (FSC) is widely used for learning from limited labeled data, yet most evaluations implicitly assume that target concepts are independent of contextual cues. In real-world settings, however, examples often appear within rich contexts, allowing models to exploit spurious correlations between foreground content and background signals. While such effects have been studied in few-shot image classification, their role in few-shot audio classification remains largely unexplored, and existing audio benchmarks offer limited control over contextual structure. We introduce SpurAud"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"many state-of-the-art few-shot methods suffer severe performance degradation when background correlations are disrupted, despite achieving similar accuracy under standard evaluation protocols. Crucially, this vulnerability persists even in large pretrained audio foundation models.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The assumption that foreground events and background environments in audio are naturally separable in a way that permits controlled, multi-level contextual shifts representative of real-world conditions.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"SpurAudio benchmark shows state-of-the-art few-shot audio classifiers suffer large performance drops when background correlations are disrupted, even in large pretrained models.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A new benchmark shows few-shot audio classifiers suffer sharp drops when background correlations are broken, even in large pretrained models.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"d97c37ac07846a13de1dc71bbb1f2b12ce06f2399983c3d2641e3daa75f5bbe1"},"source":{"id":"2605.13672","kind":"arxiv","version":1},"verdict":{"id":"02f38051-b141-425d-8fc6-cdce0236b5dd","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T20:28:37.593180Z","strongest_claim":"many state-of-the-art few-shot methods suffer severe performance degradation when background correlations are disrupted, despite achieving similar accuracy under standard evaluation protocols. Crucially, this vulnerability persists even in large pretrained audio foundation models.","one_line_summary":"SpurAudio benchmark shows state-of-the-art few-shot audio classifiers suffer large performance drops when background correlations are disrupted, even in large pretrained models.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The assumption that foreground events and background environments in audio are naturally separable in a way that permits controlled, multi-level contextual shifts representative of real-world conditions.","pith_extraction_headline":"A new benchmark shows few-shot audio classifiers suffer sharp drops when background correlations are broken, even in large pretrained models."},"references":{"count":78,"sample":[{"doi":"","year":2022,"title":"Classifying sounds in polyphonic urban sound scenes.AES E-Library","work_id":"cfc1be28-6d63-4623-baf6-fc4442ac347a","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"How robust are audio embeddings for polyphonic sound event tagging?IEEE/ACM Transactions on Audio, Speech, and Language Processing, 31:2658–2667, 2023","work_id":"90e7ab51-07fb-4902-ba84-3e5b27de3bb7","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"wav2vec 2.0: A framework for self-supervised learning of speech representations.Advances in neural information processing systems, 33:12449–12460","work_id":"03306626-15c0-4098-9991-38b25cca3d7e","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"Meta-learning with task-adaptive loss function for few-shot learning","work_id":"0b0e9e48-5e16-4e32-822a-318c4a0ec987","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2018,"title":"Meta-learning with differentiable closed-form solvers","work_id":"b613b3f1-1e74-44db-892d-defe8cdbc26a","ref_index":5,"cited_arxiv_id":"1805.08136","is_internal_anchor":true}],"resolved_work":78,"snapshot_sha256":"a1e3f169eae4fee38738389120be0cb80f62371f29075ac5a8cf68958bae5859","internal_anchors":4},"formal_canon":{"evidence_count":2,"snapshot_sha256":"07058723769ec2c3e4973861118514269643b5e31eaa24b05a53546306a812fa"},"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":"2605.13672","created_at":"2026-05-18T02:44:17.158723+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.13672v1","created_at":"2026-05-18T02:44:17.158723+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13672","created_at":"2026-05-18T02:44:17.158723+00:00"},{"alias_kind":"pith_short_12","alias_value":"ETZAOYRG4KQ3","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"ETZAOYRG4KQ3R3VF","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"ETZAOYRG","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ETZAOYRG4KQ3R3VFPO6JBRNXAE","json":"https://pith.science/pith/ETZAOYRG4KQ3R3VFPO6JBRNXAE.json","graph_json":"https://pith.science/api/pith-number/ETZAOYRG4KQ3R3VFPO6JBRNXAE/graph.json","events_json":"https://pith.science/api/pith-number/ETZAOYRG4KQ3R3VFPO6JBRNXAE/events.json","paper":"https://pith.science/paper/ETZAOYRG"},"agent_actions":{"view_html":"https://pith.science/pith/ETZAOYRG4KQ3R3VFPO6JBRNXAE","download_json":"https://pith.science/pith/ETZAOYRG4KQ3R3VFPO6JBRNXAE.json","view_paper":"https://pith.science/paper/ETZAOYRG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.13672&json=true","fetch_graph":"https://pith.science/api/pith-number/ETZAOYRG4KQ3R3VFPO6JBRNXAE/graph.json","fetch_events":"https://pith.science/api/pith-number/ETZAOYRG4KQ3R3VFPO6JBRNXAE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ETZAOYRG4KQ3R3VFPO6JBRNXAE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ETZAOYRG4KQ3R3VFPO6JBRNXAE/action/storage_attestation","attest_author":"https://pith.science/pith/ETZAOYRG4KQ3R3VFPO6JBRNXAE/action/author_attestation","sign_citation":"https://pith.science/pith/ETZAOYRG4KQ3R3VFPO6JBRNXAE/action/citation_signature","submit_replication":"https://pith.science/pith/ETZAOYRG4KQ3R3VFPO6JBRNXAE/action/replication_record"}},"created_at":"2026-05-18T02:44:17.158723+00:00","updated_at":"2026-05-18T02:44:17.158723+00:00"}