IAPO: Information-Aware Policy Optimization for Token-Efficient Reasoning

Information-aware token-level advantage assignment for accurate and token-efficient language-model reasoning.

Recommended citation: IAPO: Information-Aware Policy Optimization for Token-Efficient Reasoning. ICML 2026. Y. He, Y. Zhu, M. Shi, W. Zheng, L. Su, X. Wang, Q. Guo, J. Li
Download Paper