Ir para o conteúdo
Atlas de IA

Embeddings de texto e busca semântica

Transformar frases em vetores e buscar os mais próximos por sentido

Linguagem e conhecimentoIntermediário #09
entradaTextoTabela

O texto completo é apresentado em inglês; o título e o resumo estão traduzidos.

O QUE ESTA CAPACIDADE SIGNIFICA

Encodes a piece of text into a fixed-length vector so that semantically similar texts sit close together, then retrieves related items by nearest-neighbour lookup. Output is a ranked result list (item ids and scores), not prose. Unlike question answering it only finds possibly relevant material; it does not compose the answer.

Como é feita tecnicamente

The mainstream design is a two-tower model: queries and documents are compressed by encoders, and training pushes true pairs above random negatives, typically with contrastive learning and large batches. At query time all document vectors are pre-computed into an index and an approximate nearest-neighbour search returns the top-k in milliseconds. Sparse and dense representations are often mixed to combine keyword hits with semantic recall.

Produtos representativos

6

Organizações relacionadas

Usos típicos

  • Search over enterprise documents and code
  • The recall stage of RAG pipelines
  • Deduplication, clustering and topic discovery
  • Recommendation and similar-content entry points

Como avaliar se funciona bem

Recall@k
Whether the top-k contain all relevant documents
nDCG
Discounted cumulative gain that accounts for rank position
MRR
Mean reciprocal rank of the first relevant result

Limites e dificuldades

  • Semantic similarity is not relevance: near neighbours may merely share wording
  • Cross-domain or cross-lingual use degrades noticeably without adaptation
  • Chunking long documents severs context, and answers at chunk boundaries are easily missed

Conceitos por trás