quiz bu lëkkale ak tegtal · Diggante Niveau

ColBERT Jokkoo bu yeex

Saytu vecteur jeton ColBERT, encodage moomel sa bopp, poñ MaxSim, liggéey ci waxtu laaj, ak njëgu dencukaay bi ColBERTv2 compression di wax.

Yooni njiit yu jëm ciColbert yeexal jaxasoo
Laaj 1 de 8

naka la ColBERT di wane ab laaj wala këyit?

Xoolaatal laaj bu nekk8 laaj

Sooy def quiz bi, tontu yi ak leeral yi dañuy feeñ.

  1. naka la ColBERT di wane ab laaj wala këyit?
    • Benn vecteur buy tënk
    • Ndax baatu-caabi bu ñor bi kese
    • Vecteur yu bari yuy boole niveau token
    • Ni nataal
  2. Lu ñuy woowe fonction poñ yi ColBERT di jëfandikoo?
    • Sim
    • pooling cosinus
    • Entropi croix softmax
    • Seetug limyéer
  3. What query-time work does ColBERT avoid by precomputing document token vectors?
    • All query encoding and all index access
    • Every comparison between a query token and document token
    • The need to store document representations
    • Joint transformer encoding of every query-document pair, while query encoding and similarity search still run
  4. Luy 'late' ci 'late interaction'?
    • Royuwaay bi ñu ngi ko tàggat ci pipeline bi
    • Liggéeyukaay bi am ci digganté laaj ak token dokimaa ñu ngi koy yeexal ci jéego biy joxe poñ ginaaw enkodaas bu moom boppam
    • Laajte yi dañuy yegsi guddi gi
    • Document yi dañu leen di yokk ginaaw bi ñu leen dugalee
  5. Lan mooy jafe-jafe bi ColBERT di saafara ci am benn vecteur dense?
    • Dafa yeex lool ngir indexe
    • Mënul dox ci GPU yi
    • Dafay laaj done yuñ etikete ngir bépp laaj
    • Sooy kompresse benn passage lëmm ci benn vecteur, dafay ñàkk ay detay yu ndaw
  6. Lan la ColBERTv2 gëna yokk ci li njëkk?
    • Dina dindi bépp soxla ci samp gi
    • Dafa joge ci BERT dem ci sistem bu lalu ci sàrt
    • Yokk nañu ci kompresioŋ bu des ngir wàññi bu baax dayo index bi
    • Dafa dindi jéego MaxSim
  7. A two-token ColBERT query has best document-token similarities of 0.8 and 0.3. What MaxSim score results?
    • 0.8, retaining only the single largest value across the whole query
    • 1.1, the sum of the two per-query-token maxima
    • 0.55, averaging instead of summing the maxima
    • 0.24, multiplying the two maxima
  8. Why can a ColBERT document index require more storage than a single-vector dense index?
    • It retains multiple token-level representations for each document
    • It never precomputes any document features
    • It must store all transformer gradients for each document
    • It removes every document token before indexing