{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:2ZHJJP7TCIYAH5MDOJE4KVIJP7","short_pith_number":"pith:2ZHJJP7T","schema_version":"1.0","canonical_sha256":"d64e94bff3123003f5837249c555097fc07a89607d9db8e71f5caaf3e39ae59c","source":{"kind":"arxiv","id":"2403.08477","version":3},"attestation_state":"computed","paper":{"title":"Unleashing the Power of Meta-tuning for Few-shot Generalization Through Sparse Interpolated Experts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Jihoon Tack, Jonathan Richard Schwarz, Shengzhuang Chen, Yee Whye Teh, Ying Wei, Yunqiao Yang","submitted_at":"2024-03-13T12:46:03Z","abstract_excerpt":"Recent successes suggest that parameter-efficient fine-tuning of foundation models as the state-of-the-art method for transfer learning in vision, replacing the rich literature of alternatives such as meta-learning. In trying to harness the best of both worlds, meta-tuning introduces a subsequent optimization stage of foundation models but has so far only shown limited success and crucially tends to underperform on out-of-distribution (OOD) tasks. In this paper, we introduce Sparse MetA-Tuning (SMAT), a method inspired by sparse mixture-of-experts approaches and trained to isolate subsets of p"},"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":"2403.08477","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-03-13T12:46:03Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"6d2874d5cc171fc3351910538b4eed4b1e6183e4d81d9a6cd5b2cbbdd410f35e","abstract_canon_sha256":"88050d4bcbfdc985f4a0a756031dd25088d985986f558d710db8393bd76b3b9f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:38:37.099698Z","signature_b64":"y1PyZeKpEi+gAKB1r92l62Mjkowo3o5A9sOdxd2628w5tcFMbvyaIlyo7zvwaHRXr1+sZ/fyTqPQ5+XzLQD+Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d64e94bff3123003f5837249c555097fc07a89607d9db8e71f5caaf3e39ae59c","last_reissued_at":"2026-07-05T08:38:37.099193Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:38:37.099193Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unleashing the Power of Meta-tuning for Few-shot Generalization Through Sparse Interpolated Experts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Jihoon Tack, Jonathan Richard Schwarz, Shengzhuang Chen, Yee Whye Teh, Ying Wei, Yunqiao Yang","submitted_at":"2024-03-13T12:46:03Z","abstract_excerpt":"Recent successes suggest that parameter-efficient fine-tuning of foundation models as the state-of-the-art method for transfer learning in vision, replacing the rich literature of alternatives such as meta-learning. In trying to harness the best of both worlds, meta-tuning introduces a subsequent optimization stage of foundation models but has so far only shown limited success and crucially tends to underperform on out-of-distribution (OOD) tasks. In this paper, we introduce Sparse MetA-Tuning (SMAT), a method inspired by sparse mixture-of-experts approaches and trained to isolate subsets of p"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.08477","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/2403.08477/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":"2403.08477","created_at":"2026-07-05T08:38:37.099254+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.08477v3","created_at":"2026-07-05T08:38:37.099254+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.08477","created_at":"2026-07-05T08:38:37.099254+00:00"},{"alias_kind":"pith_short_12","alias_value":"2ZHJJP7TCIYA","created_at":"2026-07-05T08:38:37.099254+00:00"},{"alias_kind":"pith_short_16","alias_value":"2ZHJJP7TCIYAH5MD","created_at":"2026-07-05T08:38:37.099254+00:00"},{"alias_kind":"pith_short_8","alias_value":"2ZHJJP7T","created_at":"2026-07-05T08:38:37.099254+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.12597","citing_title":"Automatic Expert Discovery in LLM Upcycling via Sparse Interpolated Mixture-of-Experts","ref_index":1,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2ZHJJP7TCIYAH5MDOJE4KVIJP7","json":"https://pith.science/pith/2ZHJJP7TCIYAH5MDOJE4KVIJP7.json","graph_json":"https://pith.science/api/pith-number/2ZHJJP7TCIYAH5MDOJE4KVIJP7/graph.json","events_json":"https://pith.science/api/pith-number/2ZHJJP7TCIYAH5MDOJE4KVIJP7/events.json","paper":"https://pith.science/paper/2ZHJJP7T"},"agent_actions":{"view_html":"https://pith.science/pith/2ZHJJP7TCIYAH5MDOJE4KVIJP7","download_json":"https://pith.science/pith/2ZHJJP7TCIYAH5MDOJE4KVIJP7.json","view_paper":"https://pith.science/paper/2ZHJJP7T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.08477&json=true","fetch_graph":"https://pith.science/api/pith-number/2ZHJJP7TCIYAH5MDOJE4KVIJP7/graph.json","fetch_events":"https://pith.science/api/pith-number/2ZHJJP7TCIYAH5MDOJE4KVIJP7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2ZHJJP7TCIYAH5MDOJE4KVIJP7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2ZHJJP7TCIYAH5MDOJE4KVIJP7/action/storage_attestation","attest_author":"https://pith.science/pith/2ZHJJP7TCIYAH5MDOJE4KVIJP7/action/author_attestation","sign_citation":"https://pith.science/pith/2ZHJJP7TCIYAH5MDOJE4KVIJP7/action/citation_signature","submit_replication":"https://pith.science/pith/2ZHJJP7TCIYAH5MDOJE4KVIJP7/action/replication_record"}},"created_at":"2026-07-05T08:38:37.099254+00:00","updated_at":"2026-07-05T08:38:37.099254+00:00"}