TurnSight uses a frozen reference model plus future tool outputs to score each reasoning turn, then uses those scores to fine-tune an RL agent's credit assignment, beating existing methods on three benchmarks.
THOR:Tool-IntegratedHierar- chical Optimization via RL for Mathematical Reasoning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
TurnSight: Turn-Level Hindsight Self-Distillation for Tool-Integrated Reasoning
TurnSight uses a frozen reference model plus future tool outputs to score each reasoning turn, then uses those scores to fine-tune an RL agent's credit assignment, beating existing methods on three benchmarks.