{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ZLKOZMDDUWCGQ3XQ6HW3BJFSHA","short_pith_number":"pith:ZLKOZMDD","schema_version":"1.0","canonical_sha256":"cad4ecb063a584686ef0f1edb0a4b2380eb0a6c2404f33d981f7cf7470287b8e","source":{"kind":"arxiv","id":"2509.04606","version":1},"attestation_state":"computed","paper":{"title":"Sample-efficient Integration of New Modalities into Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.CL","authors_text":"Andr\\'e F. T. Martins, Edoardo M. Ponti, Oisin Mac Aodha, Osman Batur \\.Ince","submitted_at":"2025-09-04T18:41:59Z","abstract_excerpt":"Multimodal foundation models can process several modalities. However, since the space of possible modalities is large and evolving over time, training a model from scratch to encompass all modalities is unfeasible. Moreover, integrating a modality into a pre-existing foundation model currently requires a significant amount of paired data, which is often not available for low-resource modalities. In this paper, we introduce a method for sample-efficient modality integration (SEMI) into Large Language Models (LLMs). To this end, we devise a hypernetwork that can adapt a shared projector -- place"},"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":"2509.04606","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-04T18:41:59Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"4cd507ebd4ec32851db586f26eee06584827d611f581784dfe9f353ef32d8b47","abstract_canon_sha256":"fcd3f1ead158ce66f9ac22460a674b904eb586a2834ef4d581d3b21986d4bb8e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:05:26.137486Z","signature_b64":"+Qt30Ca28Jki2WUXltNjtfSpkmABn2Q50yg6k9zpMz8eF6U/lw6Fiy3ZUJ5JTGc360iLcT1ICcOUPoRJgoCnAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cad4ecb063a584686ef0f1edb0a4b2380eb0a6c2404f33d981f7cf7470287b8e","last_reissued_at":"2026-07-05T12:05:26.136822Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:05:26.136822Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Sample-efficient Integration of New Modalities into Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.CL","authors_text":"Andr\\'e F. T. Martins, Edoardo M. Ponti, Oisin Mac Aodha, Osman Batur \\.Ince","submitted_at":"2025-09-04T18:41:59Z","abstract_excerpt":"Multimodal foundation models can process several modalities. However, since the space of possible modalities is large and evolving over time, training a model from scratch to encompass all modalities is unfeasible. Moreover, integrating a modality into a pre-existing foundation model currently requires a significant amount of paired data, which is often not available for low-resource modalities. In this paper, we introduce a method for sample-efficient modality integration (SEMI) into Large Language Models (LLMs). To this end, we devise a hypernetwork that can adapt a shared projector -- place"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.04606","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/2509.04606/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":"2509.04606","created_at":"2026-07-05T12:05:26.136907+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.04606v1","created_at":"2026-07-05T12:05:26.136907+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.04606","created_at":"2026-07-05T12:05:26.136907+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZLKOZMDDUWCG","created_at":"2026-07-05T12:05:26.136907+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZLKOZMDDUWCGQ3XQ","created_at":"2026-07-05T12:05:26.136907+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZLKOZMDD","created_at":"2026-07-05T12:05:26.136907+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/ZLKOZMDDUWCGQ3XQ6HW3BJFSHA","json":"https://pith.science/pith/ZLKOZMDDUWCGQ3XQ6HW3BJFSHA.json","graph_json":"https://pith.science/api/pith-number/ZLKOZMDDUWCGQ3XQ6HW3BJFSHA/graph.json","events_json":"https://pith.science/api/pith-number/ZLKOZMDDUWCGQ3XQ6HW3BJFSHA/events.json","paper":"https://pith.science/paper/ZLKOZMDD"},"agent_actions":{"view_html":"https://pith.science/pith/ZLKOZMDDUWCGQ3XQ6HW3BJFSHA","download_json":"https://pith.science/pith/ZLKOZMDDUWCGQ3XQ6HW3BJFSHA.json","view_paper":"https://pith.science/paper/ZLKOZMDD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.04606&json=true","fetch_graph":"https://pith.science/api/pith-number/ZLKOZMDDUWCGQ3XQ6HW3BJFSHA/graph.json","fetch_events":"https://pith.science/api/pith-number/ZLKOZMDDUWCGQ3XQ6HW3BJFSHA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZLKOZMDDUWCGQ3XQ6HW3BJFSHA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZLKOZMDDUWCGQ3XQ6HW3BJFSHA/action/storage_attestation","attest_author":"https://pith.science/pith/ZLKOZMDDUWCGQ3XQ6HW3BJFSHA/action/author_attestation","sign_citation":"https://pith.science/pith/ZLKOZMDDUWCGQ3XQ6HW3BJFSHA/action/citation_signature","submit_replication":"https://pith.science/pith/ZLKOZMDDUWCGQ3XQ6HW3BJFSHA/action/replication_record"}},"created_at":"2026-07-05T12:05:26.136907+00:00","updated_at":"2026-07-05T12:05:26.136907+00:00"}