{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:GGQF5MMK2YGYTDRELSQKWTU6WW","short_pith_number":"pith:GGQF5MMK","schema_version":"1.0","canonical_sha256":"31a05eb18ad60d898e245ca0ab4e9eb5840951721471ca9ab10c88f6d7a06032","source":{"kind":"arxiv","id":"2505.15506","version":1},"attestation_state":"computed","paper":{"title":"Prompt Tuning Vision Language Models with Margin Regularizer for Few-Shot Learning under Distribution Shifts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Anuska Roy, Debarshi Brahma, Soma Biswas","submitted_at":"2025-05-21T13:26:56Z","abstract_excerpt":"Recently, Vision-Language foundation models like CLIP and ALIGN, which are pre-trained on large-scale data have shown remarkable zero-shot generalization to diverse datasets with different classes and even domains. In this work, we take a step further and analyze whether these models can be adapted to target datasets having very different distributions and classes compared to what these models have been trained on, using only a few labeled examples from the target dataset. In such scenarios, finetuning large pretrained models is challenging due to problems of overfitting as well as loss of gen"},"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":"2505.15506","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-21T13:26:56Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"88213beb1683aa4b82f0e6c37c34503b5efdfe93340348725be1c1c5ad9bccc7","abstract_canon_sha256":"1e2de9fc86b351dbaad6f2730b1e2b7ccbc40e6a06df11545232a9a3c099a32a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:06:46.894637Z","signature_b64":"8GBSbBC64QPpxoBfPgkMJixGmNRtbjS371SW/YKBhOrwdSh7C7X/JNraqqvT7IWxgOXJiQsgJMl6D+M8/EfPBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"31a05eb18ad60d898e245ca0ab4e9eb5840951721471ca9ab10c88f6d7a06032","last_reissued_at":"2026-07-05T11:06:46.894200Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:06:46.894200Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Prompt Tuning Vision Language Models with Margin Regularizer for Few-Shot Learning under Distribution Shifts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Anuska Roy, Debarshi Brahma, Soma Biswas","submitted_at":"2025-05-21T13:26:56Z","abstract_excerpt":"Recently, Vision-Language foundation models like CLIP and ALIGN, which are pre-trained on large-scale data have shown remarkable zero-shot generalization to diverse datasets with different classes and even domains. In this work, we take a step further and analyze whether these models can be adapted to target datasets having very different distributions and classes compared to what these models have been trained on, using only a few labeled examples from the target dataset. In such scenarios, finetuning large pretrained models is challenging due to problems of overfitting as well as loss of gen"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.15506","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/2505.15506/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":"2505.15506","created_at":"2026-07-05T11:06:46.894256+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.15506v1","created_at":"2026-07-05T11:06:46.894256+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.15506","created_at":"2026-07-05T11:06:46.894256+00:00"},{"alias_kind":"pith_short_12","alias_value":"GGQF5MMK2YGY","created_at":"2026-07-05T11:06:46.894256+00:00"},{"alias_kind":"pith_short_16","alias_value":"GGQF5MMK2YGYTDRE","created_at":"2026-07-05T11:06:46.894256+00:00"},{"alias_kind":"pith_short_8","alias_value":"GGQF5MMK","created_at":"2026-07-05T11:06:46.894256+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/GGQF5MMK2YGYTDRELSQKWTU6WW","json":"https://pith.science/pith/GGQF5MMK2YGYTDRELSQKWTU6WW.json","graph_json":"https://pith.science/api/pith-number/GGQF5MMK2YGYTDRELSQKWTU6WW/graph.json","events_json":"https://pith.science/api/pith-number/GGQF5MMK2YGYTDRELSQKWTU6WW/events.json","paper":"https://pith.science/paper/GGQF5MMK"},"agent_actions":{"view_html":"https://pith.science/pith/GGQF5MMK2YGYTDRELSQKWTU6WW","download_json":"https://pith.science/pith/GGQF5MMK2YGYTDRELSQKWTU6WW.json","view_paper":"https://pith.science/paper/GGQF5MMK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.15506&json=true","fetch_graph":"https://pith.science/api/pith-number/GGQF5MMK2YGYTDRELSQKWTU6WW/graph.json","fetch_events":"https://pith.science/api/pith-number/GGQF5MMK2YGYTDRELSQKWTU6WW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GGQF5MMK2YGYTDRELSQKWTU6WW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GGQF5MMK2YGYTDRELSQKWTU6WW/action/storage_attestation","attest_author":"https://pith.science/pith/GGQF5MMK2YGYTDRELSQKWTU6WW/action/author_attestation","sign_citation":"https://pith.science/pith/GGQF5MMK2YGYTDRELSQKWTU6WW/action/citation_signature","submit_replication":"https://pith.science/pith/GGQF5MMK2YGYTDRELSQKWTU6WW/action/replication_record"}},"created_at":"2026-07-05T11:06:46.894256+00:00","updated_at":"2026-07-05T11:06:46.894256+00:00"}