Questionário de modelos preliminares de decodificação especulativa

As respostas e explicações aparecem conforme você responde ao teste.

1. Qual é o papel principal do modelo “rascunho” na decodificação especulativa?
   • Para compactar permanentemente o modelo grande em menos parâmetros
   • Para traduzir o prompt para um idioma diferente
   • Propor rapidamente vários tokens candidatos para o grande modelo verificar
   • Para armazenar conversas anteriores para recuperação
2. Why can verifying a drafted token block accelerate target-model generation in a suitable workload?
   • Verification has no compute cost at any block size
   • A verification pass can process several positions together and amortize target work across accepted tokens
   • The draft model is guaranteed more accurate than the target
   • The target model is skipped for every accepted token
3. O que acontece com os tokens elaborados após o primeiro token rejeitado pelo modelo de destino?
   • Tudo após a primeira incompatibilidade é descartado
   • Todos são aceitos de qualquer maneira
   • Eles são enviados para um terceiro modelo
   • Eles são armazenados em cache e reutilizados no próximo turno
4. What correctness property does the standard exact speculative-sampling rejection procedure preserve under its assumptions?
   • An identical random sequence for every possible random seed
   • Guaranteed truth of every generated statement
   • The draft model's distribution regardless of the target
   • The target model's configured sampling distribution
5. Which mechanism distinguishes EAGLE's drafting approach from Medusa's additional decoding heads?
   • EAGLE performs only word-count-based lexical retrieval
   • EAGLE drafts autoregressively at the target model's feature level
   • EAGLE removes target verification from every algorithm
   • EAGLE and Medusa are the same architecture with different names
6. Why should a speculative-decoding benchmark report the draft-target pair and acceptance rate?
   • Every model pair has the same guaranteed speedup
   • Acceptance rate measures whether generated facts are true
   • Speedup depends on drafting and verification costs and how many proposed tokens are accepted
   • The target model's hardware and context length cannot affect performance
7. In exact speculative sampling, a proposed token has q equal to 0.2 and p equal to 0.1. What is its acceptance probability?
   • 0.1, because only p is used
   • 0.2, because only q is used
   • 1.0, because every proposed token is accepted
   • 0.5, from min(1, p/q)
8. What can the standard speculative-sampling algorithm emit when every token in a drafted block is accepted?
   • The accepted block plus an additional token sampled from the target
   • Only the last drafted token, discarding the accepted prefix
   • An arbitrary extra draft token without target sampling
   • No output until the draft model is retrained