{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:QQEJ4SS7B4IVEU5TD3CRAEHIPA","short_pith_number":"pith:QQEJ4SS7","schema_version":"1.0","canonical_sha256":"84089e4a5f0f115253b31ec51010e8781e448339a3cccfd7cd6fd11b95a2782d","source":{"kind":"arxiv","id":"2110.08484","version":2},"attestation_state":"computed","paper":{"title":"A Good Prompt Is Worth Millions of Parameters: Low-resource Prompt-based Learning for Vision-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Weizhu Chen, Woojeong Jin, Xiang Ren, Yelong Shen, Yu Cheng","submitted_at":"2021-10-16T06:07:59Z","abstract_excerpt":"Large pre-trained vision-language (VL) models can learn a new task with a handful of examples and generalize to a new task without fine-tuning. However, these VL models are hard to deploy for real-world applications due to their impractically huge sizes and slow inference speed. To solve this limitation, we study prompt-based low-resource learning of VL tasks with our proposed method, FewVLM, relatively smaller than recent few-shot learners. For FewVLM, we pre-train a sequence-to-sequence transformer model with prefix language modeling (PrefixLM) and masked language modeling (MaskedLM). Furthe"},"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":"2110.08484","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-10-16T06:07:59Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"34706e384e3d0b0372cdc88d2886763a4283e54cd9107c9bef9d913b22f6a146","abstract_canon_sha256":"86b975a99155d656526abda85fe07127804834581bf347d84f774da89247d35a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:05:14.015858Z","signature_b64":"22PF8KhryszVdmRTlJFR88KVHkokxG8MIjK1+wf50a7yIQzNUSZFsjHDOycxOU0iaG1o7qSI929Hk2eWT7zxCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"84089e4a5f0f115253b31ec51010e8781e448339a3cccfd7cd6fd11b95a2782d","last_reissued_at":"2026-07-05T04:05:14.015396Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:05:14.015396Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Good Prompt Is Worth Millions of Parameters: Low-resource Prompt-based Learning for Vision-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Weizhu Chen, Woojeong Jin, Xiang Ren, Yelong Shen, Yu Cheng","submitted_at":"2021-10-16T06:07:59Z","abstract_excerpt":"Large pre-trained vision-language (VL) models can learn a new task with a handful of examples and generalize to a new task without fine-tuning. However, these VL models are hard to deploy for real-world applications due to their impractically huge sizes and slow inference speed. To solve this limitation, we study prompt-based low-resource learning of VL tasks with our proposed method, FewVLM, relatively smaller than recent few-shot learners. For FewVLM, we pre-train a sequence-to-sequence transformer model with prefix language modeling (PrefixLM) and masked language modeling (MaskedLM). Furthe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.08484","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/2110.08484/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":"2110.08484","created_at":"2026-07-05T04:05:14.015455+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.08484v2","created_at":"2026-07-05T04:05:14.015455+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.08484","created_at":"2026-07-05T04:05:14.015455+00:00"},{"alias_kind":"pith_short_12","alias_value":"QQEJ4SS7B4IV","created_at":"2026-07-05T04:05:14.015455+00:00"},{"alias_kind":"pith_short_16","alias_value":"QQEJ4SS7B4IVEU5T","created_at":"2026-07-05T04:05:14.015455+00:00"},{"alias_kind":"pith_short_8","alias_value":"QQEJ4SS7","created_at":"2026-07-05T04:05:14.015455+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.14544","citing_title":"Multimodal AI for Gastrointestinal Diagnostics: Tackling VQA in MEDVQA-GI 2025","ref_index":14,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QQEJ4SS7B4IVEU5TD3CRAEHIPA","json":"https://pith.science/pith/QQEJ4SS7B4IVEU5TD3CRAEHIPA.json","graph_json":"https://pith.science/api/pith-number/QQEJ4SS7B4IVEU5TD3CRAEHIPA/graph.json","events_json":"https://pith.science/api/pith-number/QQEJ4SS7B4IVEU5TD3CRAEHIPA/events.json","paper":"https://pith.science/paper/QQEJ4SS7"},"agent_actions":{"view_html":"https://pith.science/pith/QQEJ4SS7B4IVEU5TD3CRAEHIPA","download_json":"https://pith.science/pith/QQEJ4SS7B4IVEU5TD3CRAEHIPA.json","view_paper":"https://pith.science/paper/QQEJ4SS7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.08484&json=true","fetch_graph":"https://pith.science/api/pith-number/QQEJ4SS7B4IVEU5TD3CRAEHIPA/graph.json","fetch_events":"https://pith.science/api/pith-number/QQEJ4SS7B4IVEU5TD3CRAEHIPA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QQEJ4SS7B4IVEU5TD3CRAEHIPA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QQEJ4SS7B4IVEU5TD3CRAEHIPA/action/storage_attestation","attest_author":"https://pith.science/pith/QQEJ4SS7B4IVEU5TD3CRAEHIPA/action/author_attestation","sign_citation":"https://pith.science/pith/QQEJ4SS7B4IVEU5TD3CRAEHIPA/action/citation_signature","submit_replication":"https://pith.science/pith/QQEJ4SS7B4IVEU5TD3CRAEHIPA/action/replication_record"}},"created_at":"2026-07-05T04:05:14.015455+00:00","updated_at":"2026-07-05T04:05:14.015455+00:00"}