{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ZBIFWSXZCRK25QX3ZE4WQV57YY","short_pith_number":"pith:ZBIFWSXZ","schema_version":"1.0","canonical_sha256":"c8505b4af91455aec2fbc9396857bfc63b772b1a2a5a8f40b992b9547ae02d45","source":{"kind":"arxiv","id":"2607.24354","version":1},"attestation_state":"computed","paper":{"title":"Are Prompt Optimizers Blind? Cross-Modal Visual Feedback for Automatic Prompt Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Haoyue Liu, Xiaoying Tang, Xiaoyu Ma, Ye Chen, Yuexian Zou","submitted_at":"2026-07-27T12:31:58Z","abstract_excerpt":"Automatic prompt optimization (APO) has been widely adopted to adapt vision-language models (VLMs) to downstream tasks without weight updates, yielding promising results. However, on multimodal tasks, the effectiveness of APO is fundamentally bottlenecked by a blind feedback channel: the optimizer reads the question, the prediction, and the gold answer, but never the input image on which the model failed, and therefore cannot diagnose visually grounded errors. As a remedy, we introduce Cross-Modal Visual Feedback (CMVF). CMVF incorporates (1) a failure-conditioned visual diagnosis stage, in wh"},"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":"2607.24354","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-27T12:31:58Z","cross_cats_sorted":[],"title_canon_sha256":"c337b1cb1beb64f1a46d0711724b311cf5c639e9c049e33d0260d67687bd933e","abstract_canon_sha256":"e077e947455c08faeeffd580af4ac6df67e4d70cea716257013f40d32d2322a6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-28T02:23:58.882487Z","signature_b64":"DmK7jntZhNT6RJeSVmo+LvcwpzcWjBUJeOESDm7GRKKRcy7F5eY39Xoc/pP1x6/Mj9icd2Ymfaw5ngzU68QXAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c8505b4af91455aec2fbc9396857bfc63b772b1a2a5a8f40b992b9547ae02d45","last_reissued_at":"2026-07-28T02:23:58.881587Z","signature_status":"signed_v1","first_computed_at":"2026-07-28T02:23:58.881587Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Are Prompt Optimizers Blind? Cross-Modal Visual Feedback for Automatic Prompt Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Haoyue Liu, Xiaoying Tang, Xiaoyu Ma, Ye Chen, Yuexian Zou","submitted_at":"2026-07-27T12:31:58Z","abstract_excerpt":"Automatic prompt optimization (APO) has been widely adopted to adapt vision-language models (VLMs) to downstream tasks without weight updates, yielding promising results. However, on multimodal tasks, the effectiveness of APO is fundamentally bottlenecked by a blind feedback channel: the optimizer reads the question, the prediction, and the gold answer, but never the input image on which the model failed, and therefore cannot diagnose visually grounded errors. As a remedy, we introduce Cross-Modal Visual Feedback (CMVF). CMVF incorporates (1) a failure-conditioned visual diagnosis stage, in wh"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.24354","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/2607.24354/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":"2607.24354","created_at":"2026-07-28T02:23:58.882049+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.24354v1","created_at":"2026-07-28T02:23:58.882049+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.24354","created_at":"2026-07-28T02:23:58.882049+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZBIFWSXZCRK2","created_at":"2026-07-28T02:23:58.882049+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZBIFWSXZCRK25QX3","created_at":"2026-07-28T02:23:58.882049+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZBIFWSXZ","created_at":"2026-07-28T02:23:58.882049+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/ZBIFWSXZCRK25QX3ZE4WQV57YY","json":"https://pith.science/pith/ZBIFWSXZCRK25QX3ZE4WQV57YY.json","graph_json":"https://pith.science/api/pith-number/ZBIFWSXZCRK25QX3ZE4WQV57YY/graph.json","events_json":"https://pith.science/api/pith-number/ZBIFWSXZCRK25QX3ZE4WQV57YY/events.json","paper":"https://pith.science/paper/ZBIFWSXZ"},"agent_actions":{"view_html":"https://pith.science/pith/ZBIFWSXZCRK25QX3ZE4WQV57YY","download_json":"https://pith.science/pith/ZBIFWSXZCRK25QX3ZE4WQV57YY.json","view_paper":"https://pith.science/paper/ZBIFWSXZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.24354&json=true","fetch_graph":"https://pith.science/api/pith-number/ZBIFWSXZCRK25QX3ZE4WQV57YY/graph.json","fetch_events":"https://pith.science/api/pith-number/ZBIFWSXZCRK25QX3ZE4WQV57YY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZBIFWSXZCRK25QX3ZE4WQV57YY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZBIFWSXZCRK25QX3ZE4WQV57YY/action/storage_attestation","attest_author":"https://pith.science/pith/ZBIFWSXZCRK25QX3ZE4WQV57YY/action/author_attestation","sign_citation":"https://pith.science/pith/ZBIFWSXZCRK25QX3ZE4WQV57YY/action/citation_signature","submit_replication":"https://pith.science/pith/ZBIFWSXZCRK25QX3ZE4WQV57YY/action/replication_record"}},"created_at":"2026-07-28T02:23:58.882049+00:00","updated_at":"2026-07-28T02:23:58.882049+00:00"}