{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:ETZAOYRG4KQ3R3VFPO6JBRNXAE","short_pith_number":"pith:ETZAOYRG","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"},"canonical_sha256":"24f2076226e2a1b8eea57bbc90c5b7011722e4c1f8ca587baea068a9e6471c8b","source":{"kind":"arxiv","id":"2605.13672","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.13672","created_at":"2026-05-18T02:44:17Z"},{"alias_kind":"arxiv_version","alias_value":"2605.13672v1","created_at":"2026-05-18T02:44:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13672","created_at":"2026-05-18T02:44:17Z"},{"alias_kind":"pith_short_12","alias_value":"ETZAOYRG4KQ3","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"ETZAOYRG4KQ3R3VF","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"ETZAOYRG","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:ETZAOYRG4KQ3R3VFPO6JBRNXAE","target":"record","payload":{"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"},"canonical_sha256":"24f2076226e2a1b8eea57bbc90c5b7011722e4c1f8ca587baea068a9e6471c8b","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"},"source_kind":"arxiv","source_id":"2605.13672","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T02:44:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"N8Hye/iGDHsYrJIP6IYuSGivq/N32TXDa9qATdLAypRN/oenpg9wVlXFxQ8omCKjXPqM3TA7GxGWaO0lFdqpAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-05T08:35:21.204268Z"},"content_sha256":"0c834611e0b44b6aac677ff03344476dee2ace1b30fad407955ad0a521978b4c","schema_version":"1.0","event_id":"sha256:0c834611e0b44b6aac677ff03344476dee2ace1b30fad407955ad0a521978b4c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:ETZAOYRG4KQ3R3VFPO6JBRNXAE","target":"graph","payload":{"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"},"verdict_id":"02f38051-b141-425d-8fc6-cdce0236b5dd"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T02:44:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"QLwXUhsfSHYisrx06MFbvn1aqrjMjwHTkcHcJ+Nz/MbADOZjlnqecJI2frd1bvfUbiNPVoyLiAHx+0BJfQllDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-05T08:35:21.204860Z"},"content_sha256":"1a3cba900fd1fbc4a63c219457fc0e7933d5d7c5ab8a48b8ce26298781f27264","schema_version":"1.0","event_id":"sha256:1a3cba900fd1fbc4a63c219457fc0e7933d5d7c5ab8a48b8ce26298781f27264"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ETZAOYRG4KQ3R3VFPO6JBRNXAE/bundle.json","state_url":"https://pith.science/pith/ETZAOYRG4KQ3R3VFPO6JBRNXAE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ETZAOYRG4KQ3R3VFPO6JBRNXAE/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-05T08:35:21Z","links":{"resolver":"https://pith.science/pith/ETZAOYRG4KQ3R3VFPO6JBRNXAE","bundle":"https://pith.science/pith/ETZAOYRG4KQ3R3VFPO6JBRNXAE/bundle.json","state":"https://pith.science/pith/ETZAOYRG4KQ3R3VFPO6JBRNXAE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ETZAOYRG4KQ3R3VFPO6JBRNXAE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:ETZAOYRG4KQ3R3VFPO6JBRNXAE","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"62c1ff2a536c654ae17342b7d78ac5388014e230cf35d67c6710aec7eb2d0ec2","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-13T15:32:57Z","title_canon_sha256":"831c85f859235355edf55da23da924a86619c6a93fb02c04343a993b1b33dc8a"},"schema_version":"1.0","source":{"id":"2605.13672","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.13672","created_at":"2026-05-18T02:44:17Z"},{"alias_kind":"arxiv_version","alias_value":"2605.13672v1","created_at":"2026-05-18T02:44:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13672","created_at":"2026-05-18T02:44:17Z"},{"alias_kind":"pith_short_12","alias_value":"ETZAOYRG4KQ3","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"ETZAOYRG4KQ3R3VF","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"ETZAOYRG","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:1a3cba900fd1fbc4a63c219457fc0e7933d5d7c5ab8a48b8ce26298781f27264","target":"graph","created_at":"2026-05-18T02:44:17Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A new benchmark shows few-shot audio classifiers suffer sharp drops when background correlations are broken, even in large pretrained models."}],"snapshot_sha256":"d97c37ac07846a13de1dc71bbb1f2b12ce06f2399983c3d2641e3daa75f5bbe1"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"07058723769ec2c3e4973861118514269643b5e31eaa24b05a53546306a812fa"},"paper":{"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","authors_text":"Giries Abu Ayoub, Loay Mualem, Morad Tukan","cross_cats":[],"headline":"A new benchmark shows few-shot audio classifiers suffer sharp drops when background correlations are broken, even in large pretrained models.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-13T15:32:57Z","title":"SpurAudio: A Benchmark for Studying Shortcut Learning in Few-Shot Audio Classification"},"references":{"count":78,"internal_anchors":4,"resolved_work":78,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Classifying sounds in polyphonic urban sound scenes.AES E-Library","work_id":"cfc1be28-6d63-4623-baf6-fc4442ac347a","year":2022},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"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","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"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","year":2020},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Meta-learning with task-adaptive loss function for few-shot learning","work_id":"0b0e9e48-5e16-4e32-822a-318c4a0ec987","year":2021},{"cited_arxiv_id":"1805.08136","doi":"","is_internal_anchor":true,"ref_index":5,"title":"Meta-learning with differentiable closed-form solvers","work_id":"b613b3f1-1e74-44db-892d-defe8cdbc26a","year":2018}],"snapshot_sha256":"a1e3f169eae4fee38738389120be0cb80f62371f29075ac5a8cf68958bae5859"},"source":{"id":"2605.13672","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-14T20:28:37.593180Z","id":"02f38051-b141-425d-8fc6-cdce0236b5dd","model_set":{"reader":"grok-4.3"},"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","pith_extraction_headline":"A new benchmark shows few-shot audio classifiers suffer sharp drops when background correlations are broken, even in large pretrained models.","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.","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."}},"verdict_id":"02f38051-b141-425d-8fc6-cdce0236b5dd"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:0c834611e0b44b6aac677ff03344476dee2ace1b30fad407955ad0a521978b4c","target":"record","created_at":"2026-05-18T02:44:17Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"62c1ff2a536c654ae17342b7d78ac5388014e230cf35d67c6710aec7eb2d0ec2","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-13T15:32:57Z","title_canon_sha256":"831c85f859235355edf55da23da924a86619c6a93fb02c04343a993b1b33dc8a"},"schema_version":"1.0","source":{"id":"2605.13672","kind":"arxiv","version":1}},"canonical_sha256":"24f2076226e2a1b8eea57bbc90c5b7011722e4c1f8ca587baea068a9e6471c8b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"24f2076226e2a1b8eea57bbc90c5b7011722e4c1f8ca587baea068a9e6471c8b","first_computed_at":"2026-05-18T02:44:17.158648Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T02:44:17.158648Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"WDAK5XSh0GPViOsaBsTYIODGqRzkrhicwxIPwsHs8NeJ5QUJrCG6l3BwbqZPOaEmj7cUMZvywxgbv7FHE5ZsCQ==","signature_status":"signed_v1","signed_at":"2026-05-18T02:44:17.159147Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.13672","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0c834611e0b44b6aac677ff03344476dee2ace1b30fad407955ad0a521978b4c","sha256:1a3cba900fd1fbc4a63c219457fc0e7933d5d7c5ab8a48b8ce26298781f27264"],"state_sha256":"0f1009fca11296d0e0b83bad1d301f122f892a2c70665a1e16064d2a22ebcff4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SaqvejkS174roWej83VrJ63Cq6oEvbWCrGHfnNS40Mr+sVVHG8lDPb+F/ZZZkTJHArGGCm1h71Xsr9iO6RpRCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-05T08:35:21.207281Z","bundle_sha256":"b1d4e660bc731130fb4e19b06541675f21a74fd05028f740f0394d4fb839d94c"}}