Ikibazo gihuza ibibazo · Hagati Urwego

Igihano-BERT Ikirangantego

Gerageza uburyo gushyiramo interuro bishyigikira ubushakashatsi bwimbitse, harimo Matryoshka ugereranya na cross-encoder reranking.

Inzira zijyanye no kuyoboraAmagambo ya Bert
Ikibazo 1 Bya 8

Ni ikihe kibazo cyibanze hamwe na BERT isanzwe interuro-BERT ikemura?

Reba buri kibazo8 ibibazo

Ibisubizo nibisobanuro bigaragara mugihe ufata ikibazo.

  1. Ni ikihe kibazo cyibanze hamwe na BERT isanzwe interuro-BERT ikemura?
    • Kugereranya interuro nyinshi hamwe na BERT biratinda cyane
    • BERT ntishobora gusoma interuro zirenze imwe
    • BERT ikora mucyongereza gusa
    • BERT ntishobora gutegurwa neza
  2. Ni ubuhe buryo bwubaka urusobe Sentence-BERT ikoresha?
    • Decoder imwe ya autoregressive
    • Umuyoboro wa siamese ufite ibice bibiri byo kugabana ibiro BERT
    • Umuyoboro w'amashusho
    • Umukozi-wongeyeho imbaraga
  3. Which pooling strategy did the original Sentence-BERT paper evaluate for creating sentence embeddings?
    • Max guhuza hejuru yimitwe
    • Gukoresha inyuguti yambere gusa
    • Gereranya guhuriza hejuru yikimenyetso
    • Guhitamo bisanzwe ikimenyetso kimwe
  4. Iyo interuro zimaze gushyirwamo, ni gute ubusanzwe busa?
    • Gukoresha BERT ongera kuri byombi
    • Kubara inyuguti zisangiwe
    • Kugereranya ingano ya dosiye
    • Cosine isa (ibicuruzwa byadomo) hagati ya vectors
  5. What training data did the original Sentence-BERT paper use for NLI fine-tuning?
    • Natural Language Inference examples labeled for entailment, contradiction and neutral relations
    • Only satellite images with land-cover labels
    • Only audio recordings with speaker labels
    • Only source code and bug reports
  6. In a possible semantic-retrieval workflow, what can a Sentence-BERT bi-encoder do?
    • Cross-attend to every query-candidate pair before any encoding
    • Encode a query and candidate texts separately so vector similarity can rank candidates
    • Guarantee retrieval quality for every corpus
    • Replace the need to evaluate relevance
  7. What does Matryoshka Representation Learning enable for an embedding?
    • It guarantees identical quality at every vector size
    • It converts embeddings into labeled text tokens
    • A single representation can be truncated to different dimensions to trade storage or compute for task performance
    • It removes the need for similarity comparisons
  8. In a common hybrid retrieval pipeline, what happens after a bi-encoder retrieves candidate passages?
    • The candidate text is converted into audio
    • The bi-encoder is discarded and no ranking occurs
    • The system averages all documents into one sentence
    • A cross-encoder can score query-passage pairs to rerank the candidates