{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YECUCIEQOOP4LAO45JPAZNMBE4","short_pith_number":"pith:YECUCIEQ","schema_version":"1.0","canonical_sha256":"c105412090739fc581dcea5e0cb581271f88e6225c5b431b29111ddc3602565e","source":{"kind":"arxiv","id":"2506.23856","version":1},"attestation_state":"computed","paper":{"title":"A Closer Look at Conditional Prompt Tuning for Vision-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Heng Tao Shen, Jingkuan Song, Ji Zhang, Lianli Gao, Nicu Sebe, Shihan Wu","submitted_at":"2025-06-30T13:51:20Z","abstract_excerpt":"Despite the great promise of Prompt Tuning (PT) in adapting large Vision-Language Pretrained Models (VLPMs) to downstream tasks, they often struggle to overcome the Base-New Tradeoff (BNT) dilemma: as VLPMs are better tuned to a base task, their ability to generalize to new tasks diminishes. Recent work on conditional PT addresses this problem by replacing static prompts with dynamic Visual Image Information (VII)-conditioned prompts, improving the model's generalization to new tasks to some extent. In this work, we first identify a critical issue with existing conditional PT methods: using VI"},"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":"2506.23856","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-06-30T13:51:20Z","cross_cats_sorted":[],"title_canon_sha256":"420f82bad0504a41ac9139866b58876dbd7f4edcfcbcf437a4ea0711a5c93b4a","abstract_canon_sha256":"33642c0a9596cbc42e13e2f6de691fbaf344b82ee42693b46c41edb2477ef7c2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:29:29.039773Z","signature_b64":"Ft70uo0j4d4yQZscagEwe5vkLqvi68W7dKLmtYrOH2YewUBwMBByGA1KRKYiQp+o6KGPWJ+a2HemKWzqRWh/CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c105412090739fc581dcea5e0cb581271f88e6225c5b431b29111ddc3602565e","last_reissued_at":"2026-07-05T11:29:29.039281Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:29:29.039281Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Closer Look at Conditional Prompt Tuning for Vision-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Heng Tao Shen, Jingkuan Song, Ji Zhang, Lianli Gao, Nicu Sebe, Shihan Wu","submitted_at":"2025-06-30T13:51:20Z","abstract_excerpt":"Despite the great promise of Prompt Tuning (PT) in adapting large Vision-Language Pretrained Models (VLPMs) to downstream tasks, they often struggle to overcome the Base-New Tradeoff (BNT) dilemma: as VLPMs are better tuned to a base task, their ability to generalize to new tasks diminishes. Recent work on conditional PT addresses this problem by replacing static prompts with dynamic Visual Image Information (VII)-conditioned prompts, improving the model's generalization to new tasks to some extent. In this work, we first identify a critical issue with existing conditional PT methods: using VI"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.23856","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/2506.23856/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":"2506.23856","created_at":"2026-07-05T11:29:29.039341+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.23856v1","created_at":"2026-07-05T11:29:29.039341+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.23856","created_at":"2026-07-05T11:29:29.039341+00:00"},{"alias_kind":"pith_short_12","alias_value":"YECUCIEQOOP4","created_at":"2026-07-05T11:29:29.039341+00:00"},{"alias_kind":"pith_short_16","alias_value":"YECUCIEQOOP4LAO4","created_at":"2026-07-05T11:29:29.039341+00:00"},{"alias_kind":"pith_short_8","alias_value":"YECUCIEQ","created_at":"2026-07-05T11:29:29.039341+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24248","citing_title":"M^2C-EvDet: Multi-Domain Multi-Order Cross-Modal Knowledge Distillation for Event-based Object Detection","ref_index":15,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YECUCIEQOOP4LAO45JPAZNMBE4","json":"https://pith.science/pith/YECUCIEQOOP4LAO45JPAZNMBE4.json","graph_json":"https://pith.science/api/pith-number/YECUCIEQOOP4LAO45JPAZNMBE4/graph.json","events_json":"https://pith.science/api/pith-number/YECUCIEQOOP4LAO45JPAZNMBE4/events.json","paper":"https://pith.science/paper/YECUCIEQ"},"agent_actions":{"view_html":"https://pith.science/pith/YECUCIEQOOP4LAO45JPAZNMBE4","download_json":"https://pith.science/pith/YECUCIEQOOP4LAO45JPAZNMBE4.json","view_paper":"https://pith.science/paper/YECUCIEQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.23856&json=true","fetch_graph":"https://pith.science/api/pith-number/YECUCIEQOOP4LAO45JPAZNMBE4/graph.json","fetch_events":"https://pith.science/api/pith-number/YECUCIEQOOP4LAO45JPAZNMBE4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YECUCIEQOOP4LAO45JPAZNMBE4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YECUCIEQOOP4LAO45JPAZNMBE4/action/storage_attestation","attest_author":"https://pith.science/pith/YECUCIEQOOP4LAO45JPAZNMBE4/action/author_attestation","sign_citation":"https://pith.science/pith/YECUCIEQOOP4LAO45JPAZNMBE4/action/citation_signature","submit_replication":"https://pith.science/pith/YECUCIEQOOP4LAO45JPAZNMBE4/action/replication_record"}},"created_at":"2026-07-05T11:29:29.039341+00:00","updated_at":"2026-07-05T11:29:29.039341+00:00"}