Nucleus uye Top-k Sampling Quiz

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

1. Ko top-k sampling inorambidza sei sarudzo dzetokeni?
   • Inochengeta chete k yepamusoro-inogoneka tokens, uye masampuli pakati pawo
   • Inochengeta tokens kusvika mukana wekuwedzeredza wapfuura k
   • Inogara ichisarudza chiratidzo chimwe chete chingangoitika
   • Iyo inobvisa k zvishoma zvingangoitika zviratidzo chete
2. Chii chinotsanangudza 'nucleus' munucleus (top-p) sampling?
   • Iwo k anonyanya kutaurwa mazwi mukudzidziswa data
   • Iyo diki seti yematokeni ayo mikana inosvika painokwana p
   • Iyo imwechete yepamusoro-inogoneka chiratidzo
   • Tokeni dzese dzine mukana uri pamusoro pe50%
3. Ndeipi yakakosha mukana wetop-p pamusoro pepamusoro-k?
   • Nguva dzose inokurumidza kuverenga
   • Haimbobudisi zvakangoitika
   • Huwandu hwema tokens emumiriri anochinjika kune chivimbo chemuenzaniso
   • Iyo inofuratira zvingangoitika zvechiratidzo
4. Chii chinoita tembiricha parameter isati yatorwa sampling?
   • Inoisa iyo yakanyanya kutevedzana kureba
   • Inosarudza mutauro wekubuda
   • Inoverenga hukuru hwemashoko
   • Inodzoreredza mukana wekugovera kuve wakapinza kana kupfava
5. What result did the nucleus-sampling paper report for its tested open-ended generation tasks?
   • It improved diversity without sacrificing fluency and coherence in the reported comparisons
   • It guaranteed factual correctness in every output
   • It eliminated all sampling randomness
   • It improved translation quality on every benchmark
6. For a greedy decoding baseline with no token sampling, how should generation be configured?
   • Disable sampling and use greedy decoding to select the highest-probability token at each step
   • Use high temperature and a large nucleus threshold
   • Randomize the temperature at every step
   • Keep sampling enabled but increase top-p
7. After top-k truncation removes tokens, what happens to the remaining sampling probabilities?
   • They remain assigned their original total mass below one
   • They are replaced with equal probabilities
   • They are renormalized over the retained top-k tokens
   • They are converted into beam-search scores
8. What did Holtzman et al. report about likelihood-maximizing decoding for open-ended text?
   • It always yields the most diverse human-like passages
   • Inobvisa kudiwa kwemuenzaniso wemutauro
   • It guarantees factual correctness
   • It can produce bland and repetitive text despite strong likelihood scores