Neexal bu ñu boole ci Quiz RLHF

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

1. Ci normalisasioŋ bu neexal buñ boole, ak lan lañuy méngale ak neexal bu tontu bu nekk?
   • Moyenne bi ak écart standard bi ci grupu prompt bi
   • ab constant global bu fiks
   • Neexal bi am ci jamono yi weesu
   • lim parametru model bi
2. Ban algorithm moo gëna méngoo ak normalisasioŋ bu neexal buñ boole?
   • Wàcci degrade stokastik
   • GRPO (Optimisation ci wàllu politik relatif)
   • K-dafay tekki boole
   • Tasaaro ci ginaaw
3. Ban cër ci PPO standard la GRPO di dindi?
   • Royuwaayu neexal bi
   • Reseau neuronal bi lëmm
   • Reseau valeur bu tàqaloo (critique)
   • Done yiñ joxe ci tàggat yaram
4. Lan mooy dall tontu bu neexal bi nekk ci suufu li grupu bi mëna am?
   • Dafay am njariñ lu baax te dañu koy dooleel
   • Dañu ko efaase ci done yi
   • Neexal bi dafa yokk ñaari yoon
   • Dafay am njariñ lu baaxul te dañu koy tere
5. What does within-prompt reward normalization provide to a GRPO update?
   • A relative, scaled advantage for each response compared with others for the same prompt
   • A guarantee that all prompt difficulty differences disappear
   • A higher learning rate
   • A way to make every response receive the same reward
6. What happens when every response in a prompt group receives the same reward?
   • Every response gets a positive advantage
   • Centering gives no relative advantage; zero spread also requires numerical handling
   • The group mean doubles the reward
   • The model stops needing a policy objective
7. Which PPO-style update element does GRPO retain?
   • A separately learned value network
   • K-means clustering over tokens
   • A clipped policy objective
   • A language-model tokenizer
8. What can a KL penalty relative to a reference policy help constrain in GRPO?
   • The number of prompts in each group
   • The length of the tokenizer vocabulary
   • The resolution of generated images
   • How far the updated policy drifts from the reference policy