quiz bu lëkkale ak tegtal · Diggante Niveau

Quiz bi gëna néew doole

Saytu ni Least-to-Most di xaajalee laaj yu jafe yi ak jëfandikoo ay pexe yu toppalante ngir gëna yombal generalisation bu yomb ba bu jafe.

Yooni njiit yu jëm ciLi gëna ndaw ba gëna am doole
Laaj 1 de 8

Lan mooy ñaari wàll yi gëna am solo ci ñaaxtu bu gëna ndaw ba gëna am solo?

Xoolaatal laaj bu nekk8 laaj

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

  1. Lan mooy ñaari wàll yi gëna am solo ci ñaaxtu bu gëna ndaw ba gëna am solo?
    • Xaajaleel jafe-jafe bi ay laaj yu ndaw, ba noppi nga def leen ñu toppalante
    • Taggat nga defar ko bu baax
    • Enkode te nga dekode
    • Wutal tontu yu bari te nga wote
  2. Lan moo wuutale Least-to-Most ak li ñuy woowe chaîne de pensée?
    • Jëfandikuwul benn misaal
    • Dafay leer ci jafe-jafe bi ba noppi dugal tontu bu ndaw bu nekk ci jéego bi ci topp
    • Dafay laaj nga defar model bi bu baax
    • Ci nataal kese lay dox
  3. What kind of task does the SCAN benchmark use to test compositional generalization?
    • Classifying satellite images of fishing vessels
    • Predicting fuel use from ship weather data
    • Interpreting natural-language commands that combine primitive actions
    • Aligning speech frames with handwritten characters
  4. Lan moo waral ñu toppalante laaj yu ndaw yi am solo?
    • Du noonu; komànd bi dafay mujjee
    • Laaj yi ci topp ñoo wara njëkka ñëw
    • Komànd du am njeexital ci format bi
    • Bépp laaj bu ndaw dafa wara aju ci laaj yiñ jota joxe
  5. naka lañuy jëfandikoo tontu yu digg yi ci wàllu saafara?
    • Dañu leen boole ci muy tekki suko defee jéego yi ci topp mëna tabax ci kaw
    • Jéego bu nekk dañu leen di sànni
    • Dañu leen di yónnee ci beneen model
    • Dañu leen di boole ci benn nimero
  6. Lan mooy risk bu mag bi ci chaine Least-to-Most?
    • Mënul jëfandikoo misaal
    • Tontu bu teel bu juum mën na yàq bépp jéego bu wàcci
    • Benn làkk kese lay dox
    • Dafay laaj cluster GPU saa yu nekk
  7. What accuracy did the paper report for code-davinci-002 on the SCAN length split using Least-to-Most with 14 exemplars?
    • About 16%
    • 50% ci dëgg
    • At least 99% on the reported setup
    • Zero, because no demonstrations were used
  8. Besides SCAN, which task areas did the Least-to-Most paper also evaluate?
    • Wildlife re-identification and acoustic classification
    • Ship routing and port scheduling
    • Image outpainting and watermark detection
    • Symbolic manipulation and math reasoning