{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:OABOTDEFI7EGASIMOQUU7TPDDR","short_pith_number":"pith:OABOTDEF","schema_version":"1.0","canonical_sha256":"7002e98c8547c860490c74294fcde31c67ea0550fefbd33159ec68a0fd78b68f","source":{"kind":"arxiv","id":"2303.02506","version":3},"attestation_state":"computed","paper":{"title":"Prismer: A Vision-Language Model with Multi-Task Experts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Anima Anandkumar, Chaowei Xiao, Edward Johns, Linxi Fan, Shikun Liu, Zhiding Yu","submitted_at":"2023-03-04T21:22:47Z","abstract_excerpt":"Recent vision-language models have shown impressive multi-modal generation capabilities. However, typically they require training huge models on massive datasets. As a more scalable alternative, we introduce Prismer, a data- and parameter-efficient vision-language model that leverages an ensemble of task-specific experts. Prismer only requires training of a small number of components, with the majority of network weights inherited from multiple readily-available, pre-trained experts, and kept frozen during training. By leveraging experts from a wide range of domains, we show Prismer can effici"},"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":"2303.02506","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-03-04T21:22:47Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"d870775460b4e94243b67f7f9c0a102aa6c4ec6d49d8fda999d8549374a0d6be","abstract_canon_sha256":"ce4742aa7a8c96dd16d77b8ebf666433e8860aa470669621338fca42e4e07ef6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:35:18.190552Z","signature_b64":"uWAIqEkAgTEAF9msazI5wagntCaHZugzYmFVKmfSOeXZeqFNSGNYK7dI9ww2mVteloInKfGB4fxx67hsZdmxCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7002e98c8547c860490c74294fcde31c67ea0550fefbd33159ec68a0fd78b68f","last_reissued_at":"2026-07-05T07:35:18.190075Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:35:18.190075Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Prismer: A Vision-Language Model with Multi-Task Experts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Anima Anandkumar, Chaowei Xiao, Edward Johns, Linxi Fan, Shikun Liu, Zhiding Yu","submitted_at":"2023-03-04T21:22:47Z","abstract_excerpt":"Recent vision-language models have shown impressive multi-modal generation capabilities. However, typically they require training huge models on massive datasets. As a more scalable alternative, we introduce Prismer, a data- and parameter-efficient vision-language model that leverages an ensemble of task-specific experts. Prismer only requires training of a small number of components, with the majority of network weights inherited from multiple readily-available, pre-trained experts, and kept frozen during training. By leveraging experts from a wide range of domains, we show Prismer can effici"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.02506","kind":"arxiv","version":3},"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/2303.02506/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":"2303.02506","created_at":"2026-07-05T07:35:18.190130+00:00"},{"alias_kind":"arxiv_version","alias_value":"2303.02506v3","created_at":"2026-07-05T07:35:18.190130+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.02506","created_at":"2026-07-05T07:35:18.190130+00:00"},{"alias_kind":"pith_short_12","alias_value":"OABOTDEFI7EG","created_at":"2026-07-05T07:35:18.190130+00:00"},{"alias_kind":"pith_short_16","alias_value":"OABOTDEFI7EGASIM","created_at":"2026-07-05T07:35:18.190130+00:00"},{"alias_kind":"pith_short_8","alias_value":"OABOTDEF","created_at":"2026-07-05T07:35:18.190130+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2309.16797","citing_title":"Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution","ref_index":213,"is_internal_anchor":false},{"citing_arxiv_id":"2311.03079","citing_title":"CogVLM: Visual Expert for Pretrained Language Models","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2305.03726","citing_title":"Otter: A Multi-Modal Model with In-Context Instruction Tuning","ref_index":58,"is_internal_anchor":false},{"citing_arxiv_id":"2305.16291","citing_title":"Voyager: An Open-Ended Embodied Agent with Large Language Models","ref_index":58,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OABOTDEFI7EGASIMOQUU7TPDDR","json":"https://pith.science/pith/OABOTDEFI7EGASIMOQUU7TPDDR.json","graph_json":"https://pith.science/api/pith-number/OABOTDEFI7EGASIMOQUU7TPDDR/graph.json","events_json":"https://pith.science/api/pith-number/OABOTDEFI7EGASIMOQUU7TPDDR/events.json","paper":"https://pith.science/paper/OABOTDEF"},"agent_actions":{"view_html":"https://pith.science/pith/OABOTDEFI7EGASIMOQUU7TPDDR","download_json":"https://pith.science/pith/OABOTDEFI7EGASIMOQUU7TPDDR.json","view_paper":"https://pith.science/paper/OABOTDEF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2303.02506&json=true","fetch_graph":"https://pith.science/api/pith-number/OABOTDEFI7EGASIMOQUU7TPDDR/graph.json","fetch_events":"https://pith.science/api/pith-number/OABOTDEFI7EGASIMOQUU7TPDDR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OABOTDEFI7EGASIMOQUU7TPDDR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OABOTDEFI7EGASIMOQUU7TPDDR/action/storage_attestation","attest_author":"https://pith.science/pith/OABOTDEFI7EGASIMOQUU7TPDDR/action/author_attestation","sign_citation":"https://pith.science/pith/OABOTDEFI7EGASIMOQUU7TPDDR/action/citation_signature","submit_replication":"https://pith.science/pith/OABOTDEFI7EGASIMOQUU7TPDDR/action/replication_record"}},"created_at":"2026-07-05T07:35:18.190130+00:00","updated_at":"2026-07-05T07:35:18.190130+00:00"}