Learning What Reinforcement Learning Can’t: Interleaved Online Fine-Tuning for Hardest Questions

Recommended citation: L Ma, H Liang, M Qiang, L Tang, X Ma, ZH Wong, J Niu, C Shen, R He, et al. (2026). "Learning What Reinforcement Learning Can't: Interleaved Online Fine-Tuning for Hardest Questions." ICLR 2026.