DiSP stratifies queries by difficulty using random trial estimates, trains a router and level-specific judges, then applies budgeted stop-on-acceptance selection to improve ICL accuracy and speed on classification tasks.
The Llama 3 Herd of Models , 2024
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Easier to Judge than to Find: Predicting In-Context Learning Success for Demonstration Selection
DiSP stratifies queries by difficulty using random trial estimates, trains a router and level-specific judges, then applies budgeted stop-on-acceptance selection to improve ICL accuracy and speed on classification tasks.