{
  "altLabels": [
    "Multimodal Search",
    "Multimedia Retrieval",
    "Multimedia Information Retrieval"
  ],
  "definition": "Retrieval involving information needs, indexed items, representations, or results that span more than one data modality, such as text, images, audio, video, tables, or structured metadata.",
  "derivedRelations": [
    {
      "target": "cross-modal-retrieval",
      "type": "narrower"
    },
    {
      "target": "image-retrieval",
      "type": "narrower"
    },
    {
      "note": "TRECVID is an evaluation campaign for video analysis and retrieval; multimodal retrieval is a broader retrieval task family.",
      "target": "trecvid",
      "type": "notEquivalentTo"
    },
    {
      "target": "semantic-search",
      "type": "related"
    },
    {
      "target": "tip-of-the-tongue-search",
      "type": "related"
    },
    {
      "target": "trecvid",
      "type": "related"
    }
  ],
  "distinctions": [
    {
      "reason": "Multimodal retrieval is the broader family involving multiple modalities; cross-modal retrieval specifically crosses a modality boundary between query and retrieved items.",
      "statement": "Multimodal Retrieval is not Cross-Modal Retrieval.",
      "target": "cross-modal-retrieval",
      "targetLabel": "Cross-Modal Retrieval",
      "targetUri": "https://id.searchplex.net/cross-modal-retrieval/"
    },
    {
      "reason": "Image retrieval targets images; multimodal retrieval is broader and may involve images, text, video, audio, tables, or combinations of modalities.",
      "statement": "Multimodal Retrieval is not Image Retrieval.",
      "target": "image-retrieval",
      "targetLabel": "Image Retrieval",
      "targetUri": "https://id.searchplex.net/image-retrieval/"
    },
    {
      "reason": "Multimodal retrieval is about data modalities; multi-vector retrieval is about representing a query or item with multiple vectors.",
      "statement": "Multimodal Retrieval is not Multi-Vector Retrieval.",
      "target": "multi-vector-retrieval",
      "targetLabel": "Multi-Vector Retrieval",
      "targetUri": "https://id.searchplex.net/multi-vector-retrieval/"
    },
    {
      "reason": "Multimodal retrieval may be used inside multimodal RAG systems, but RAG is a retrieval-augmented generation pattern.",
      "statement": "Multimodal Retrieval is not Retrieval-Augmented Generation.",
      "target": "rag",
      "targetLabel": "Retrieval-Augmented Generation",
      "targetUri": "https://id.searchplex.net/rag/"
    }
  ],
  "id": "multimodal-retrieval",
  "kind": "Task",
  "license": "https://creativecommons.org/publicdomain/zero/1.0/",
  "links": [
    {
      "label": "Ask in Any Modality",
      "type": "canonicalSurvey",
      "url": "https://arxiv.org/abs/2502.08826"
    }
  ],
  "modifiedAt": "2026-09-14T06:43:46Z",
  "prefLabel": "Multimodal Retrieval",
  "publisher": {
    "name": "Searchplex",
    "url": "https://searchplex.net/"
  },
  "relations": [
    {
      "target": "retrieval",
      "type": "broader"
    },
    {
      "note": "Multimodal retrieval is the broader family involving multiple modalities; cross-modal retrieval specifically crosses a modality boundary between query and retrieved items.",
      "target": "cross-modal-retrieval",
      "type": "notEquivalentTo"
    },
    {
      "note": "Image retrieval targets images; multimodal retrieval is broader and may involve images, text, video, audio, tables, or combinations of modalities.",
      "target": "image-retrieval",
      "type": "notEquivalentTo"
    },
    {
      "note": "Multimodal retrieval is about data modalities; multi-vector retrieval is about representing a query or item with multiple vectors.",
      "target": "multi-vector-retrieval",
      "type": "notEquivalentTo"
    },
    {
      "note": "Multimodal retrieval may be used inside multimodal RAG systems, but RAG is a retrieval-augmented generation pattern.",
      "target": "rag",
      "type": "notEquivalentTo"
    },
    {
      "target": "clef",
      "type": "related"
    },
    {
      "target": "multi-vector-retrieval",
      "type": "related"
    },
    {
      "target": "rag",
      "type": "related"
    }
  ],
  "representations": {
    "html": "https://id.searchplex.net/multimodal-retrieval/",
    "json": "https://id.searchplex.net/multimodal-retrieval.json",
    "jsonld": "https://id.searchplex.net/multimodal-retrieval.jsonld"
  },
  "scheme": "https://id.searchplex.net/scheme/",
  "scopeNote": "Multimodal retrieval is a broad family. It includes settings where multiple modalities are searched together, represented jointly, or used in query and result interpretation. It is distinct from cross-modal retrieval, where the query and target results are in different modalities, and from image retrieval, where the retrieved objects are images.",
  "status": "published",
  "terms": [
    {
      "community": "vendor",
      "label": "Multimodal Vector Search",
      "usage": "nearSynonym"
    },
    {
      "community": "industry",
      "label": "Neural Multimodal Search",
      "usage": "nearSynonym"
    }
  ],
  "uri": "https://id.searchplex.net/multimodal-retrieval/"
}
