A distillation framework that composes solutions from a tool-augmented agent and a text-reasoning teacher trains a 7B model to dynamically choose between code execution and verbal reasoning, improving math benchmark accuracy.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Agentic-R1: Distilled Dual-Strategy Reasoning
A distillation framework that composes solutions from a tool-augmented agent and a text-reasoning teacher trains a 7B model to dynamically choose between code execution and verbal reasoning, improving math benchmark accuracy.