{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:GHYYKOU3XMGQLTTW6ACAUHZBVY","short_pith_number":"pith:GHYYKOU3","schema_version":"1.0","canonical_sha256":"31f1853a9bbb0d05ce76f0040a1f21ae29128ded58b3f208967447f4caf15a40","source":{"kind":"arxiv","id":"2404.06212","version":1},"attestation_state":"computed","paper":{"title":"OmniFusion Technical Report","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Andrey Kuznetsov, Anton Razzhigaev, Denis Dimitrov, Elizaveta Goncharova, Irina Abdullaeva, Ivan Oseledets, Matvey Mikhalchuk, Matvey Skripkin, Maxim Kurkin","submitted_at":"2024-04-09T11:00:19Z","abstract_excerpt":"Last year, multimodal architectures served up a revolution in AI-based approaches and solutions, extending the capabilities of large language models (LLM). We propose an \\textit{OmniFusion} model based on a pretrained LLM and adapters for visual modality. We evaluated and compared several architecture design principles for better text and visual data coupling: MLP and transformer adapters, various CLIP ViT-based encoders (SigLIP, InternVIT, etc.), and their fusing approach, image encoding method (whole image or tiles encoding) and two 7B LLMs (the proprietary one and open-source Mistral). Expe"},"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":"2404.06212","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-04-09T11:00:19Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"4893efa0233a060e0eef6b1fb0eea42026e4fae404163d33e0ee9500eeaad1eb","abstract_canon_sha256":"e426540062eefb76a0b47f1b5f148243e43f28d0edc7303bd27aeac932077fc1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:06:05.812404Z","signature_b64":"x6+/3wr6FPj6NIMcAdzyPZp3MfA4tobcRuP4vRxTcb2WmMXAWC6CCwBcmRwDeo8shkQDVmR7M7QgCagU1LYlCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"31f1853a9bbb0d05ce76f0040a1f21ae29128ded58b3f208967447f4caf15a40","last_reissued_at":"2026-07-05T08:06:05.811933Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:06:05.811933Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"OmniFusion Technical Report","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Andrey Kuznetsov, Anton Razzhigaev, Denis Dimitrov, Elizaveta Goncharova, Irina Abdullaeva, Ivan Oseledets, Matvey Mikhalchuk, Matvey Skripkin, Maxim Kurkin","submitted_at":"2024-04-09T11:00:19Z","abstract_excerpt":"Last year, multimodal architectures served up a revolution in AI-based approaches and solutions, extending the capabilities of large language models (LLM). We propose an \\textit{OmniFusion} model based on a pretrained LLM and adapters for visual modality. We evaluated and compared several architecture design principles for better text and visual data coupling: MLP and transformer adapters, various CLIP ViT-based encoders (SigLIP, InternVIT, etc.), and their fusing approach, image encoding method (whole image or tiles encoding) and two 7B LLMs (the proprietary one and open-source Mistral). Expe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.06212","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/2404.06212/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":"2404.06212","created_at":"2026-07-05T08:06:05.811991+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.06212v1","created_at":"2026-07-05T08:06:05.811991+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.06212","created_at":"2026-07-05T08:06:05.811991+00:00"},{"alias_kind":"pith_short_12","alias_value":"GHYYKOU3XMGQ","created_at":"2026-07-05T08:06:05.811991+00:00"},{"alias_kind":"pith_short_16","alias_value":"GHYYKOU3XMGQLTTW","created_at":"2026-07-05T08:06:05.811991+00:00"},{"alias_kind":"pith_short_8","alias_value":"GHYYKOU3","created_at":"2026-07-05T08:06:05.811991+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.11531","citing_title":"Addressing Hallucinations in Language Models with Knowledge Graph Embeddings as an Additional Modality","ref_index":12,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GHYYKOU3XMGQLTTW6ACAUHZBVY","json":"https://pith.science/pith/GHYYKOU3XMGQLTTW6ACAUHZBVY.json","graph_json":"https://pith.science/api/pith-number/GHYYKOU3XMGQLTTW6ACAUHZBVY/graph.json","events_json":"https://pith.science/api/pith-number/GHYYKOU3XMGQLTTW6ACAUHZBVY/events.json","paper":"https://pith.science/paper/GHYYKOU3"},"agent_actions":{"view_html":"https://pith.science/pith/GHYYKOU3XMGQLTTW6ACAUHZBVY","download_json":"https://pith.science/pith/GHYYKOU3XMGQLTTW6ACAUHZBVY.json","view_paper":"https://pith.science/paper/GHYYKOU3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.06212&json=true","fetch_graph":"https://pith.science/api/pith-number/GHYYKOU3XMGQLTTW6ACAUHZBVY/graph.json","fetch_events":"https://pith.science/api/pith-number/GHYYKOU3XMGQLTTW6ACAUHZBVY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GHYYKOU3XMGQLTTW6ACAUHZBVY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GHYYKOU3XMGQLTTW6ACAUHZBVY/action/storage_attestation","attest_author":"https://pith.science/pith/GHYYKOU3XMGQLTTW6ACAUHZBVY/action/author_attestation","sign_citation":"https://pith.science/pith/GHYYKOU3XMGQLTTW6ACAUHZBVY/action/citation_signature","submit_replication":"https://pith.science/pith/GHYYKOU3XMGQLTTW6ACAUHZBVY/action/replication_record"}},"created_at":"2026-07-05T08:06:05.811991+00:00","updated_at":"2026-07-05T08:06:05.811991+00:00"}