{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:CWNRZTEBKIWWDJ7DFBHOWNIMQJ","short_pith_number":"pith:CWNRZTEB","schema_version":"1.0","canonical_sha256":"159b1ccc81522d61a7e3284eeb350c826a737a5c9fe83d06fc085ec705a0615e","source":{"kind":"arxiv","id":"2409.07825","version":4},"attestation_state":"computed","paper":{"title":"Deep Multimodal Learning with Missing Modality: A Survey","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Multimodal deep learning models can maintain performance when some input types are missing by using dedicated robustness techniques.","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Gustavo Carneiro, Hsiang-Ting Chen, Hu Wang, Renjie Wu","submitted_at":"2024-09-12T08:15:39Z","abstract_excerpt":"During multimodal model training and testing, certain data modalities may be absent due to sensor limitations, cost constraints, privacy concerns, or data loss, negatively affecting performance. Multimodal learning techniques designed to handle missing modalities can mitigate this by ensuring model robustness even when some modalities are unavailable. This survey reviews recent progress in Multimodal Learning with Missing Modality (MLMM), focusing on deep learning methods. It provides the first comprehensive survey that covers the motivation and distinctions between MLMM and standard multimoda"},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2409.07825","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-09-12T08:15:39Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"c74c2012fa9a4f98b57808185868664a2192599e11bc9e8c24c759ceeb8565ab","abstract_canon_sha256":"6fa6031b7706d3a4ad81ff54065cf302441618c58152a992913a1f70df345060"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:12.837208Z","signature_b64":"geWHQEOHVtLrpQQIJK2dCvsHj07dwUxc4SXHoR9eCmm6oi/2m7o7GNX1Iee1MoW+nQFisqGk/ehg3PUY8yP6Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"159b1ccc81522d61a7e3284eeb350c826a737a5c9fe83d06fc085ec705a0615e","last_reissued_at":"2026-05-17T23:38:12.836498Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:12.836498Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Deep Multimodal Learning with Missing Modality: A Survey","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Multimodal deep learning models can maintain performance when some input types are missing by using dedicated robustness techniques.","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Gustavo Carneiro, Hsiang-Ting Chen, Hu Wang, Renjie Wu","submitted_at":"2024-09-12T08:15:39Z","abstract_excerpt":"During multimodal model training and testing, certain data modalities may be absent due to sensor limitations, cost constraints, privacy concerns, or data loss, negatively affecting performance. Multimodal learning techniques designed to handle missing modalities can mitigate this by ensuring model robustness even when some modalities are unavailable. This survey reviews recent progress in Multimodal Learning with Missing Modality (MLMM), focusing on deep learning methods. It provides the first comprehensive survey that covers the motivation and distinctions between MLMM and standard multimoda"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"It provides the first comprehensive survey that covers the motivation and distinctions between MLMM and standard multimodal learning setups, followed by a detailed analysis of current methods, applications, and datasets, concluding with challenges and future directions.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The assumption that the body of literature selected for review is sufficiently complete and representative of the current state of deep multimodal learning with missing modalities without major omissions of recent or niche contributions.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"This survey provides the first comprehensive overview of deep multimodal learning methods designed to remain robust when some input modalities are absent.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Multimodal deep learning models can maintain performance when some input types are missing by using dedicated robustness techniques.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"326ee4fc199e8f58663c8527cb84f61b03e71bcf4c66fe2da827e7543c868484"},"source":{"id":"2409.07825","kind":"arxiv","version":4},"verdict":{"id":"1dc0160e-90b8-45dc-b1f6-5abd166717d4","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-17T22:22:32.472970Z","strongest_claim":"It provides the first comprehensive survey that covers the motivation and distinctions between MLMM and standard multimodal learning setups, followed by a detailed analysis of current methods, applications, and datasets, concluding with challenges and future directions.","one_line_summary":"This survey provides the first comprehensive overview of deep multimodal learning methods designed to remain robust when some input modalities are absent.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The assumption that the body of literature selected for review is sufficiently complete and representative of the current state of deep multimodal learning with missing modalities without major omissions of recent or niche contributions.","pith_extraction_headline":"Multimodal deep learning models can maintain performance when some input types are missing by using dedicated robustness techniques."},"references":{"count":83,"sample":[{"doi":"","year":null,"title":"Medical image segmentation on mri images with missing modalities: A review.arXiv preprint arXiv:2203.06217,","work_id":"26faf823-1334-4867-8301-3bb958c01481","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2026,"title":"Dealing with the effects of sensor displacement in wearable activity recognition.Sensors, 14(6):9995–10023,","work_id":"6810563a-21df-403c-b6f4-e590859e0568","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.24432/c5c59f","year":null,"title":"Rohan Bavishi, Erich Elsen, Curtis Hawthorne, Maxwell Nye, Augustus Odena, Arushi Somani, and Sağnak Taşırlar","work_id":"860663e0-e63a-4ea2-ab8e-b9bfe266c7f8","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2018,"title":"Overcoming missing and incomplete modalities with generative adversarial networks for building footprint segmentation","work_id":"5b3e97e9-8576-4e4d-b7a6-65aeee8208ef","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Sparks of Artificial General Intelligence: Early experiments with GPT-4","work_id":"a23cfe92-7f7c-424b-98d4-b386a83002fb","ref_index":5,"cited_arxiv_id":"2303.12712","is_internal_anchor":true}],"resolved_work":83,"snapshot_sha256":"30af0375504c607d74e1e01c1575a5521f06e0d252606d8dffcadf47ffbf1e26","internal_anchors":9},"formal_canon":{"evidence_count":2,"snapshot_sha256":"8d11bfc3c28d94f260dcfd0683a5f078ad3982d7da9be81e3b58c599e9aa11e8"},"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":"2409.07825","created_at":"2026-05-17T23:38:12.836619+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.07825v4","created_at":"2026-05-17T23:38:12.836619+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.07825","created_at":"2026-05-17T23:38:12.836619+00:00"},{"alias_kind":"pith_short_12","alias_value":"CWNRZTEBKIWW","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"CWNRZTEBKIWWDJ7D","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"CWNRZTEB","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":17,"internal_anchor_count":17,"sample":[{"citing_arxiv_id":"2511.12034","citing_title":"Calibrated Multimodal Representation Learning with Missing Modalities","ref_index":54,"is_internal_anchor":true},{"citing_arxiv_id":"2512.22991","citing_title":"Fusion or Confusion? Multimodal Complexity Is Not All You Need","ref_index":51,"is_internal_anchor":true},{"citing_arxiv_id":"2601.22853","citing_title":"Inference-Time Dynamic Modality Selection for Incomplete Multimodal Classification","ref_index":5,"is_internal_anchor":true},{"citing_arxiv_id":"2602.16161","citing_title":"Emotion Collider: Dual Hyperbolic Mirror Manifolds for Sentiment Recovery via Anti Emotion Reflection","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2602.16197","citing_title":"ModalImmune: Immunity Driven Unlearning via Self Destructive Training","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12031","citing_title":"Resilient Vision-Tabular Multimodal Learning under Modality Missingness","ref_index":5,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08302","citing_title":"SGC-RML: A reliable and interpretable longitudinal assessment for PD in real-world DNS","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2604.22212","citing_title":"Multimodal Diffusion to Mutually Enhance Polarized Light and Low Resolution EBSD Data","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2604.22885","citing_title":"Federated Cross-Modal Retrieval with Missing Modalities via Semantic Routing and Adapter Personalization","ref_index":38,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06086","citing_title":"LARGO: Low-Rank Hypernetwork for Handling Missing Modalities","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2605.00670","citing_title":"Robust Multimodal Recommendation via Graph Retrieval-Enhanced Modality Completion","ref_index":45,"is_internal_anchor":true},{"citing_arxiv_id":"2604.10695","citing_title":"Retrieving to Recover: Towards Incomplete Audio-Visual Question Answering via Semantic-consistent Purification","ref_index":2,"is_internal_anchor":true},{"citing_arxiv_id":"2604.09711","citing_title":"Head-wise Modality Specialization within MLLMs for Robust Fake News Detection under Missing Modality","ref_index":39,"is_internal_anchor":true},{"citing_arxiv_id":"2604.05584","citing_title":"Purify-then-Align: Towards Robust Human Sensing under Modality Missing with Knowledge Distillation from Noisy Multimodal Teacher","ref_index":40,"is_internal_anchor":true},{"citing_arxiv_id":"2604.05558","citing_title":"Evaluation Before Generation: A Paradigm for Robust Multimodal Sentiment Analysis with Missing Modalities","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2604.17030","citing_title":"Conditional Evidence Reconstruction and Decomposition for Interpretable Multimodal Diagnosis","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2604.20283","citing_title":"Multi-Perspective Evidence Synthesis and Reasoning for Unsupervised Multimodal Entity Linking","ref_index":51,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CWNRZTEBKIWWDJ7DFBHOWNIMQJ","json":"https://pith.science/pith/CWNRZTEBKIWWDJ7DFBHOWNIMQJ.json","graph_json":"https://pith.science/api/pith-number/CWNRZTEBKIWWDJ7DFBHOWNIMQJ/graph.json","events_json":"https://pith.science/api/pith-number/CWNRZTEBKIWWDJ7DFBHOWNIMQJ/events.json","paper":"https://pith.science/paper/CWNRZTEB"},"agent_actions":{"view_html":"https://pith.science/pith/CWNRZTEBKIWWDJ7DFBHOWNIMQJ","download_json":"https://pith.science/pith/CWNRZTEBKIWWDJ7DFBHOWNIMQJ.json","view_paper":"https://pith.science/paper/CWNRZTEB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.07825&json=true","fetch_graph":"https://pith.science/api/pith-number/CWNRZTEBKIWWDJ7DFBHOWNIMQJ/graph.json","fetch_events":"https://pith.science/api/pith-number/CWNRZTEBKIWWDJ7DFBHOWNIMQJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CWNRZTEBKIWWDJ7DFBHOWNIMQJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CWNRZTEBKIWWDJ7DFBHOWNIMQJ/action/storage_attestation","attest_author":"https://pith.science/pith/CWNRZTEBKIWWDJ7DFBHOWNIMQJ/action/author_attestation","sign_citation":"https://pith.science/pith/CWNRZTEBKIWWDJ7DFBHOWNIMQJ/action/citation_signature","submit_replication":"https://pith.science/pith/CWNRZTEBKIWWDJ7DFBHOWNIMQJ/action/replication_record"}},"created_at":"2026-05-17T23:38:12.836619+00:00","updated_at":"2026-05-17T23:38:12.836619+00:00"}