Guide-yakabatana mibvunzo · Pakati Nhanho

BM25 uye Lexical Retrieval Quiz

Ongorora BM25 izwi rarity, frequency saturation, zvinyorwa-kureba zvidzoreso, inverted indexes, uye basa rayo mumigodhi yakaoma negative kune dense retrievers.

Nzira dzekutungamira dzinoenderanaBm25 And Lexical Retrieval
Mubvunzo 1 ye 8

BM25 inonyanya kushandisa chii kuisa magwaro?

Tarisa mubvunzo wese8 mibvunzo

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

  1. BM25 inonyanya kushandisa chii kuisa magwaro?
    • Dense semantic embeddings
    • Term frequency, inverse document frequency, uye kureba kwegwaro
    • Zuva rekuburitswa kwegwaro rega rega
    • Mushandisi kubaya-kuburikidza nenhoroondo chete
  2. Ndeipi basa rinoitwa inverse gwaro frequency (IDF) muBM25?
    • Inosimudzira mazwi akajairika senge 'the'
    • Inoyera kuti gwaro rega rega rakareba sei
    • Inopa zvisingawanzo, mamwe mazwi anosiyanisa huremu hwepamusoro
    • Inoverenga nhamba yose yezvinyorwa chete
  3. Sei BM25 ichishandisa gwaro-kureba normalization (iyo b parameter)?
    • Saka kuti magwaro marefu haafarirwe zvisizvo nekungova nemamwe mazwi
    • Kuita kuti gwaro rega rega rive nehurefu hwakafanana
    • Kudzima magwaro mapfupi
    • Kuturikira magwaro mumutauro wakajairika
  4. Chinhu chakakosha cheBM25's term-frequency kubata ndeyekuti:
    • Inowedzera mutsara nekusingaperi nekudzokorora kwega kwega
    • Inofuratira frequency yetemu zvachose
    • Inopeta zvibodzwa zvezwi rega rega rinodzokororwa
    • Saturates, saka kudzokorora kwakawedzera kunowedzera kudzikira kukosha
  5. Ndechipi chimiro che data chinoita kuti BM25 ikurumidze pachiyero?
    • Inverted index mepu mazwi kune zvinyorwa anazvo
    • A neural network
    • Mutsara umwechete wakarongwa wezvinyorwa zvese
    • Dhatabhesi regirafu rezvinyorwa
  6. Mumazuva ano ekutsvaga masisitimu, BM25 inowanzo shandiswa ku:
    • Tsiva ese neural modhi zvachose
    • Shandira seyekukurumidza nhanho yekutanga kudzoreredza pamberi peneural reranker, kana kusanganiswa neakakora zvibodzwa.
    • Gadzira zvinyorwa zvitsva
    • Encrypt index yekutsvaga
  7. In the standard BM25 formula, what effect does setting b to zero have?
    • It removes all inverse-document-frequency weights
    • It makes every matching term contribute zero
    • It changes the index from lexical terms to dense vectors
    • It disables document-length normalization in the term-frequency denominator
  8. When BM25 is used to mine DPR training negatives, what makes a retrieved nonanswer passage a hard negative?
    • It has no words in common with any document
    • It is automatically the positive passage because its lexical score is high
    • It looks relevant through term overlap but does not support the question's answer
    • It contains no readable text and is never retrieved