Many-shot CoT-ICL improves when demonstrations are ordered for smooth conceptual progression, with CDS delivering up to 5.42 percentage-point gains on math tasks using 64 examples.
Transformers learn in-context by gradient descent
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CL 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
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.
citing papers explorer
-
Many-Shot CoT-ICL: Making In-Context Learning Truly Learn
Many-shot CoT-ICL improves when demonstrations are ordered for smooth conceptual progression, with CDS delivering up to 5.42 percentage-point gains on math tasks using 64 examples.
-
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.