{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:A5GUN3II27HLZAPW6VU6CJ7QRA","short_pith_number":"pith:A5GUN3II","schema_version":"1.0","canonical_sha256":"074d46ed08d7cebc81f6f569e127f0882303b7c4562a86f9ca1bb662089c2bd1","source":{"kind":"arxiv","id":"2406.03146","version":2},"attestation_state":"computed","paper":{"title":"Tiny models from tiny data: Textual and null-text inversion for few-shot distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Erik Landolsi, Fredrik Kahl","submitted_at":"2024-06-05T11:01:42Z","abstract_excerpt":"Few-shot learning deals with problems such as image classification using very few training examples. Recent vision foundation models show excellent few-shot transfer abilities, but are large and slow at inference. Using knowledge distillation, the capabilities of high-performing but slow models can be transferred to tiny, efficient models. However, common distillation methods require a large set of unlabeled data, which is not available in the few-shot setting. To overcome this lack of data, there has been a recent interest in using synthetic data. We expand on this line of research by present"},"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":"2406.03146","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-06-05T11:01:42Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"0be4737ce9a2012bfb8624df1806a4ca968fc1d88c53d2a6086ed678ea6642a5","abstract_canon_sha256":"d6874704ab03e7a66d30ac373c9b5c484c911a19bf9375306e6cde4d81f4df2d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:35:33.429110Z","signature_b64":"LKMoVmxFmBUempOz3i1C1TMiBCFnbuc/p/RAeWj92iGd/7w/xFICe1bd9AF2pJMcpjEcEV8VObegH5ds0HG0AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"074d46ed08d7cebc81f6f569e127f0882303b7c4562a86f9ca1bb662089c2bd1","last_reissued_at":"2026-07-05T10:35:33.428588Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:35:33.428588Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Tiny models from tiny data: Textual and null-text inversion for few-shot distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Erik Landolsi, Fredrik Kahl","submitted_at":"2024-06-05T11:01:42Z","abstract_excerpt":"Few-shot learning deals with problems such as image classification using very few training examples. Recent vision foundation models show excellent few-shot transfer abilities, but are large and slow at inference. Using knowledge distillation, the capabilities of high-performing but slow models can be transferred to tiny, efficient models. However, common distillation methods require a large set of unlabeled data, which is not available in the few-shot setting. To overcome this lack of data, there has been a recent interest in using synthetic data. We expand on this line of research by present"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.03146","kind":"arxiv","version":2},"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/2406.03146/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":"2406.03146","created_at":"2026-07-05T10:35:33.428657+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.03146v2","created_at":"2026-07-05T10:35:33.428657+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.03146","created_at":"2026-07-05T10:35:33.428657+00:00"},{"alias_kind":"pith_short_12","alias_value":"A5GUN3II27HL","created_at":"2026-07-05T10:35:33.428657+00:00"},{"alias_kind":"pith_short_16","alias_value":"A5GUN3II27HLZAPW","created_at":"2026-07-05T10:35:33.428657+00:00"},{"alias_kind":"pith_short_8","alias_value":"A5GUN3II","created_at":"2026-07-05T10:35:33.428657+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.07481","citing_title":"Addressing degeneracies in latent interpolation for diffusion models","ref_index":12,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/A5GUN3II27HLZAPW6VU6CJ7QRA","json":"https://pith.science/pith/A5GUN3II27HLZAPW6VU6CJ7QRA.json","graph_json":"https://pith.science/api/pith-number/A5GUN3II27HLZAPW6VU6CJ7QRA/graph.json","events_json":"https://pith.science/api/pith-number/A5GUN3II27HLZAPW6VU6CJ7QRA/events.json","paper":"https://pith.science/paper/A5GUN3II"},"agent_actions":{"view_html":"https://pith.science/pith/A5GUN3II27HLZAPW6VU6CJ7QRA","download_json":"https://pith.science/pith/A5GUN3II27HLZAPW6VU6CJ7QRA.json","view_paper":"https://pith.science/paper/A5GUN3II","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.03146&json=true","fetch_graph":"https://pith.science/api/pith-number/A5GUN3II27HLZAPW6VU6CJ7QRA/graph.json","fetch_events":"https://pith.science/api/pith-number/A5GUN3II27HLZAPW6VU6CJ7QRA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A5GUN3II27HLZAPW6VU6CJ7QRA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A5GUN3II27HLZAPW6VU6CJ7QRA/action/storage_attestation","attest_author":"https://pith.science/pith/A5GUN3II27HLZAPW6VU6CJ7QRA/action/author_attestation","sign_citation":"https://pith.science/pith/A5GUN3II27HLZAPW6VU6CJ7QRA/action/citation_signature","submit_replication":"https://pith.science/pith/A5GUN3II27HLZAPW6VU6CJ7QRA/action/replication_record"}},"created_at":"2026-07-05T10:35:33.428657+00:00","updated_at":"2026-07-05T10:35:33.428657+00:00"}