{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:J5HMXKZDEFDMSAGCSMTYKVWUIU","short_pith_number":"pith:J5HMXKZD","schema_version":"1.0","canonical_sha256":"4f4ecbab232146c900c293278556d4450a8f57515656ac1db1e361b4ce4ddaa4","source":{"kind":"arxiv","id":"2210.07179","version":2},"attestation_state":"computed","paper":{"title":"MAPL: Parameter-Efficient Adaptation of Unimodal Pre-Trained Models for Vision-Language Few-Shot Prompting","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Aida Nematzadeh, Aishwarya Agrawal, Oscar Ma\\~nas, Pau Rodriguez, Saba Ahmadi, Yash Goyal","submitted_at":"2022-10-13T17:02:23Z","abstract_excerpt":"Large pre-trained models have proved to be remarkable zero- and (prompt-based) few-shot learners in unimodal vision and language tasks. We propose MAPL, a simple and parameter-efficient method that reuses frozen pre-trained unimodal models and leverages their strong generalization capabilities in multimodal vision-language (VL) settings. MAPL learns a lightweight mapping between the representation spaces of unimodal models using aligned image-text data, and can generalize to unseen VL tasks from just a few in-context examples. The small number of trainable parameters makes MAPL effective at lo"},"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":"2210.07179","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-10-13T17:02:23Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"ba4de3b6c2f059032dfd7a210c7edc8aae121bd7761d2abaa88d0539f0209847","abstract_canon_sha256":"5a4a404f5637005467db44feee9e28262da14bb4253a4c7ddbb7d014c3520b3d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:51:27.620887Z","signature_b64":"NGrIHMGLNLn/ADAl3ZzRu8TX170gvhrtACxukalTkN1QeUM/J7YEsz51/t4XeTGG07feiYBvvI9AzQBHSBxdBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4f4ecbab232146c900c293278556d4450a8f57515656ac1db1e361b4ce4ddaa4","last_reissued_at":"2026-07-05T05:51:27.620366Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:51:27.620366Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MAPL: Parameter-Efficient Adaptation of Unimodal Pre-Trained Models for Vision-Language Few-Shot Prompting","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Aida Nematzadeh, Aishwarya Agrawal, Oscar Ma\\~nas, Pau Rodriguez, Saba Ahmadi, Yash Goyal","submitted_at":"2022-10-13T17:02:23Z","abstract_excerpt":"Large pre-trained models have proved to be remarkable zero- and (prompt-based) few-shot learners in unimodal vision and language tasks. We propose MAPL, a simple and parameter-efficient method that reuses frozen pre-trained unimodal models and leverages their strong generalization capabilities in multimodal vision-language (VL) settings. MAPL learns a lightweight mapping between the representation spaces of unimodal models using aligned image-text data, and can generalize to unseen VL tasks from just a few in-context examples. The small number of trainable parameters makes MAPL effective at lo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.07179","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/2210.07179/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":"2210.07179","created_at":"2026-07-05T05:51:27.620430+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.07179v2","created_at":"2026-07-05T05:51:27.620430+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.07179","created_at":"2026-07-05T05:51:27.620430+00:00"},{"alias_kind":"pith_short_12","alias_value":"J5HMXKZDEFDM","created_at":"2026-07-05T05:51:27.620430+00:00"},{"alias_kind":"pith_short_16","alias_value":"J5HMXKZDEFDMSAGC","created_at":"2026-07-05T05:51:27.620430+00:00"},{"alias_kind":"pith_short_8","alias_value":"J5HMXKZD","created_at":"2026-07-05T05:51:27.620430+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.12633","citing_title":"ECA: Efficient Continual Alignment for Open-Ended Image-to-Text Generation","ref_index":63,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/J5HMXKZDEFDMSAGCSMTYKVWUIU","json":"https://pith.science/pith/J5HMXKZDEFDMSAGCSMTYKVWUIU.json","graph_json":"https://pith.science/api/pith-number/J5HMXKZDEFDMSAGCSMTYKVWUIU/graph.json","events_json":"https://pith.science/api/pith-number/J5HMXKZDEFDMSAGCSMTYKVWUIU/events.json","paper":"https://pith.science/paper/J5HMXKZD"},"agent_actions":{"view_html":"https://pith.science/pith/J5HMXKZDEFDMSAGCSMTYKVWUIU","download_json":"https://pith.science/pith/J5HMXKZDEFDMSAGCSMTYKVWUIU.json","view_paper":"https://pith.science/paper/J5HMXKZD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.07179&json=true","fetch_graph":"https://pith.science/api/pith-number/J5HMXKZDEFDMSAGCSMTYKVWUIU/graph.json","fetch_events":"https://pith.science/api/pith-number/J5HMXKZDEFDMSAGCSMTYKVWUIU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/J5HMXKZDEFDMSAGCSMTYKVWUIU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/J5HMXKZDEFDMSAGCSMTYKVWUIU/action/storage_attestation","attest_author":"https://pith.science/pith/J5HMXKZDEFDMSAGCSMTYKVWUIU/action/author_attestation","sign_citation":"https://pith.science/pith/J5HMXKZDEFDMSAGCSMTYKVWUIU/action/citation_signature","submit_replication":"https://pith.science/pith/J5HMXKZDEFDMSAGCSMTYKVWUIU/action/replication_record"}},"created_at":"2026-07-05T05:51:27.620430+00:00","updated_at":"2026-07-05T05:51:27.620430+00:00"}