alleryStart learning →Many-shot ICL can perform comparably to
ASupervised fine-tuning on the same task data
BReinforcement learning with explicit reward shaping
CA separately pretrained model with twice the params
DRetrieval augmentation over the training corpus
Answer & Solution
Correct answer: A. Supervised fine-tuning on the same task data
Section 4.3 shows many-shot ICL is competitive with supervised fine-tuning on the same data — but without weight updates and with one model snapshot serving many tasks.
Keep practising Many Shot in Context Learning
Browse all Many Shot in Context Learning questions for Claude, or explore more Aifoundations topics.
Related questions
Rule on these many shot claimsThe implication for prompt design when shots are abundant isLong-context scaling laws (next-token loss)Many-shot ICL can overcomeThe order of examples in many-shot promptsAbstractive summarization on XSumMachine translation many-shot resultsThe maximum number of shots evaluated is approximately