Introduces IBPO, a counterfactual credit assignment method that turns sparse terminal rewards into process-level advantage estimates for more stable LLM reasoning training.
Selec- tive reflection-tuning: Student-selected data recycling for llm instruction-tuning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Reducing Credit Assignment Variance via Counterfactual Reasoning Paths
Introduces IBPO, a counterfactual credit assignment method that turns sparse terminal rewards into process-level advantage estimates for more stable LLM reasoning training.