{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:OBAVFK3JQWRRRSPR3XOZIAES3S","short_pith_number":"pith:OBAVFK3J","schema_version":"1.0","canonical_sha256":"704152ab6985a318c9f1dddd940092dc89a6a8a6f21ec1949041e0689bf00e0a","source":{"kind":"arxiv","id":"2507.13089","version":1},"attestation_state":"computed","paper":{"title":"GLAD: Generalizable Tuning for Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jianzhuang Liu, Pengfei Wang, Shifeng Chen, Yuqi Peng","submitted_at":"2025-07-17T12:58:15Z","abstract_excerpt":"Pre-trained vision-language models, such as CLIP, show impressive zero-shot recognition ability and can be easily transferred to specific downstream tasks via prompt tuning, even with limited training data. However, existing prompt tuning methods face two main challenges: (1) In few-shot scenarios, data scarcity often leads to overfitting, making the model sensitive to changes in the input domain. (2) To mitigate overfitting, these methods typically rely on complex task-specific model architectures and sensitive hyperparameter tuning, severely restricting their general applicability. To addres"},"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":"2507.13089","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-07-17T12:58:15Z","cross_cats_sorted":[],"title_canon_sha256":"5bfe696ca976e05ca62e59aba11ada61012cc9d2ee186347d0ddecf668cf7c8f","abstract_canon_sha256":"1fbffae566bdd5b20f8f64e5b4cc2dd408eb32c6b91969ef9b80012805925c85"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:38:54.550467Z","signature_b64":"3BePfvNUT/5QIGeo55zb80la+LWMlNK5V/kvUnTfiBcoUjj7BlQfW7e20g/IS2ky8JMJ+Nt3yZN/bs7RIq5hBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"704152ab6985a318c9f1dddd940092dc89a6a8a6f21ec1949041e0689bf00e0a","last_reissued_at":"2026-07-05T11:38:54.549999Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:38:54.549999Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GLAD: Generalizable Tuning for Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jianzhuang Liu, Pengfei Wang, Shifeng Chen, Yuqi Peng","submitted_at":"2025-07-17T12:58:15Z","abstract_excerpt":"Pre-trained vision-language models, such as CLIP, show impressive zero-shot recognition ability and can be easily transferred to specific downstream tasks via prompt tuning, even with limited training data. However, existing prompt tuning methods face two main challenges: (1) In few-shot scenarios, data scarcity often leads to overfitting, making the model sensitive to changes in the input domain. (2) To mitigate overfitting, these methods typically rely on complex task-specific model architectures and sensitive hyperparameter tuning, severely restricting their general applicability. To addres"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.13089","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/2507.13089/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":"2507.13089","created_at":"2026-07-05T11:38:54.550050+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.13089v1","created_at":"2026-07-05T11:38:54.550050+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.13089","created_at":"2026-07-05T11:38:54.550050+00:00"},{"alias_kind":"pith_short_12","alias_value":"OBAVFK3JQWRR","created_at":"2026-07-05T11:38:54.550050+00:00"},{"alias_kind":"pith_short_16","alias_value":"OBAVFK3JQWRRRSPR","created_at":"2026-07-05T11:38:54.550050+00:00"},{"alias_kind":"pith_short_8","alias_value":"OBAVFK3J","created_at":"2026-07-05T11:38:54.550050+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.14262","citing_title":"GUI-Perturbed: Domain Randomization Reveals Systematic Brittleness in GUI Grounding Models","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OBAVFK3JQWRRRSPR3XOZIAES3S","json":"https://pith.science/pith/OBAVFK3JQWRRRSPR3XOZIAES3S.json","graph_json":"https://pith.science/api/pith-number/OBAVFK3JQWRRRSPR3XOZIAES3S/graph.json","events_json":"https://pith.science/api/pith-number/OBAVFK3JQWRRRSPR3XOZIAES3S/events.json","paper":"https://pith.science/paper/OBAVFK3J"},"agent_actions":{"view_html":"https://pith.science/pith/OBAVFK3JQWRRRSPR3XOZIAES3S","download_json":"https://pith.science/pith/OBAVFK3JQWRRRSPR3XOZIAES3S.json","view_paper":"https://pith.science/paper/OBAVFK3J","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.13089&json=true","fetch_graph":"https://pith.science/api/pith-number/OBAVFK3JQWRRRSPR3XOZIAES3S/graph.json","fetch_events":"https://pith.science/api/pith-number/OBAVFK3JQWRRRSPR3XOZIAES3S/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OBAVFK3JQWRRRSPR3XOZIAES3S/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OBAVFK3JQWRRRSPR3XOZIAES3S/action/storage_attestation","attest_author":"https://pith.science/pith/OBAVFK3JQWRRRSPR3XOZIAES3S/action/author_attestation","sign_citation":"https://pith.science/pith/OBAVFK3JQWRRRSPR3XOZIAES3S/action/citation_signature","submit_replication":"https://pith.science/pith/OBAVFK3JQWRRRSPR3XOZIAES3S/action/replication_record"}},"created_at":"2026-07-05T11:38:54.550050+00:00","updated_at":"2026-07-05T11:38:54.550050+00:00"}