{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:OSJHYCOFV7W6MIMA42SLDMJVCQ","short_pith_number":"pith:OSJHYCOF","schema_version":"1.0","canonical_sha256":"74927c09c5afede62180e6a4b1b13514166f5fb81d349ac21ae13e6af6dc25db","source":{"kind":"arxiv","id":"2411.15099","version":1},"attestation_state":"computed","paper":{"title":"Context-Aware Multimodal Pretraining","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Dima Damen, Ivana Bala\\v{z}evi\\'c, Karsten Roth, Olivier J. H\\'enaff, Zeynep Akata","submitted_at":"2024-11-22T17:55:39Z","abstract_excerpt":"Large-scale multimodal representation learning successfully optimizes for zero-shot transfer at test time. Yet the standard pretraining paradigm (contrastive learning on large amounts of image-text data) does not explicitly encourage representations to support few-shot adaptation. In this work, we propose a simple, but carefully designed extension to multimodal pretraining which enables representations to accommodate additional context. Using this objective, we show that vision-language models can be trained to exhibit significantly increased few-shot adaptation: across 21 downstream tasks, we"},"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":"2411.15099","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-11-22T17:55:39Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"a712d0e994cb7c0b90caa9edb18a277ae1faae59cd525175e15240fc825bd12b","abstract_canon_sha256":"86f3038a7ff4a8a1bebeb3daf24f3bbe846a8079364832e903f2d34a2d39aaa0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:39:16.358610Z","signature_b64":"RMogpaCrfIZLRW0XNdYFGQJ5+fZNEvlEzqo0U15yz5P7h7bcv7MJI+H9y91XMo6K9wCxZCToNL6wp3/VVrYRDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"74927c09c5afede62180e6a4b1b13514166f5fb81d349ac21ae13e6af6dc25db","last_reissued_at":"2026-07-05T09:39:16.358187Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:39:16.358187Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Context-Aware Multimodal Pretraining","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Dima Damen, Ivana Bala\\v{z}evi\\'c, Karsten Roth, Olivier J. H\\'enaff, Zeynep Akata","submitted_at":"2024-11-22T17:55:39Z","abstract_excerpt":"Large-scale multimodal representation learning successfully optimizes for zero-shot transfer at test time. Yet the standard pretraining paradigm (contrastive learning on large amounts of image-text data) does not explicitly encourage representations to support few-shot adaptation. In this work, we propose a simple, but carefully designed extension to multimodal pretraining which enables representations to accommodate additional context. Using this objective, we show that vision-language models can be trained to exhibit significantly increased few-shot adaptation: across 21 downstream tasks, we"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.15099","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/2411.15099/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":"2411.15099","created_at":"2026-07-05T09:39:16.358237+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.15099v1","created_at":"2026-07-05T09:39:16.358237+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.15099","created_at":"2026-07-05T09:39:16.358237+00:00"},{"alias_kind":"pith_short_12","alias_value":"OSJHYCOFV7W6","created_at":"2026-07-05T09:39:16.358237+00:00"},{"alias_kind":"pith_short_16","alias_value":"OSJHYCOFV7W6MIMA","created_at":"2026-07-05T09:39:16.358237+00:00"},{"alias_kind":"pith_short_8","alias_value":"OSJHYCOF","created_at":"2026-07-05T09:39:16.358237+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.06712","citing_title":"How to Merge Your Multimodal Models Over Time?","ref_index":66,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OSJHYCOFV7W6MIMA42SLDMJVCQ","json":"https://pith.science/pith/OSJHYCOFV7W6MIMA42SLDMJVCQ.json","graph_json":"https://pith.science/api/pith-number/OSJHYCOFV7W6MIMA42SLDMJVCQ/graph.json","events_json":"https://pith.science/api/pith-number/OSJHYCOFV7W6MIMA42SLDMJVCQ/events.json","paper":"https://pith.science/paper/OSJHYCOF"},"agent_actions":{"view_html":"https://pith.science/pith/OSJHYCOFV7W6MIMA42SLDMJVCQ","download_json":"https://pith.science/pith/OSJHYCOFV7W6MIMA42SLDMJVCQ.json","view_paper":"https://pith.science/paper/OSJHYCOF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.15099&json=true","fetch_graph":"https://pith.science/api/pith-number/OSJHYCOFV7W6MIMA42SLDMJVCQ/graph.json","fetch_events":"https://pith.science/api/pith-number/OSJHYCOFV7W6MIMA42SLDMJVCQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OSJHYCOFV7W6MIMA42SLDMJVCQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OSJHYCOFV7W6MIMA42SLDMJVCQ/action/storage_attestation","attest_author":"https://pith.science/pith/OSJHYCOFV7W6MIMA42SLDMJVCQ/action/author_attestation","sign_citation":"https://pith.science/pith/OSJHYCOFV7W6MIMA42SLDMJVCQ/action/citation_signature","submit_replication":"https://pith.science/pith/OSJHYCOFV7W6MIMA42SLDMJVCQ/action/replication_record"}},"created_at":"2026-07-05T09:39:16.358237+00:00","updated_at":"2026-07-05T09:39:16.358237+00:00"}