@misc{indiciae4fad704fd355,
  title = {Balancing the Reasoning Load: Difficulty-Differentiated Policy Optimization with Length Redistribution for Efficient and Robust Reinforcement Learning},
  author = {Yinan Xia and Haotian Zhang and Huiming Wang},
  year = {2026},
  url = {https://arxiv.org/abs/2603.18533},
  note = {Source identifier: 2603.18533}
}