Matryoshka Matalaadda Kediska Ku Xidhan

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

1. Waa maxay hantida asaasiga ah ee ku-xidhka Matryoshka?
   • Trained prefixes can serve as representations at multiple lengths
   • Cabir kastaa waa isku si oo muhiim ah oo nidaamku macno ma leh
   • Waxa kaliya loo isticmaali karaa hal dherer oo go'an
   • Waa in dib loogu dhejiyaa cabbir kasta oo bartilmaameed ah
2. Sidee loo tababaraa qaabka Matryoshka si loo gaaro tan?
   • Iyadoo loo tababarayo qaab gaar ah oo cabbir kasta ah
   • Iyadoo khasaare lagu soo koobay cabbirro horgaleyaal badan oo buul leh
   • Adigoo codsanaya PCA ka dib tababarka caadiga ah
   • Iyada oo si aan kala sooc lahayn hoos loogu dhigayo cabbirrada wakhtiga tijaabada oo keliya
3. Maxaad samaynaysaa fikradda si aad u hesho ku-xidhid yar?
   • Dib u tabobar qaabka meel eber
   • Samee shabakad cadaadis gaar ah
   • U jar cabbirka k u horreeya oo dib u habeyn
   • Isku celcelis dhammaan cabbirrada wadajirka ah
4. Which documented API lets a caller request fewer dimensions for text-embedding-3 output?
   • The dimensions parameter
   • A PCA flag in the Matryoshka paper
   • A FastText n-gram option
   • A GloVe context-window setting
5. Waa maxay faa'iidada asaasiga ah ee aasaasiga ah ee ku-xidhka Matryoshka?
   • Fewer coordinates may reduce storage and comparison work; benchmark retrieval quality and cost
   • Waxay meesha ka saarayaan baahida loo qabo vector kasta
   • Waxay dammaanad qaadayaan saxnaanta qumman ee dhammaan cabbirrada
   • Waxay kaliya u shaqeeyaan sawirada
6. Ku-xidhka Matryoshka, xagee ku uruursan tahay xogta ugu takoorka badan?
   • Dhammaadka, cabbirrada daba-socda
   • Marka ugu horeysa, cabbirrada hogaaminaya
   • U faafi si isku mid ah iyada oo aan la dalban
   • Kaliya cabbirada dhexe
7. Which dimensionality schedule is an example of nested Matryoshka prefixes in the guide?
   • Only one fixed size of 512 dimensions
   • A decreasing schedule from 2048 to 1 dimension with separate models
   • 8, 16, 32, and progressively larger prefixes up to 2048 dimensions
   • A sequence of unrelated sizes requiring new model weights each time
8. How does truncating a Matryoshka embedding differ from applying a separate PCA model?
   • It requires retraining a different encoder for every dimension
   • It replaces the original vector with a random projection
   • It can only be done while the model is being trained
   • A shorter prefix can be sliced and renormalized from the same trained embedding without a separate projection model