We propose TraceRL, a trajectory-aware reinforcement learning method for diffusion language models, which demonstrates the best performance among RL approaches for DLMs. We also introduce a ...

Figure 1. FIPO vs. baselines on AIME 2024. FIPO shows that pure RL training alone can outperform reproduced pure-RL baselines such as DAPO and DeepSeek-R1-Zero-32B, surpass o1-mini, and produce ...

Abstract: With the advancement of marine resource development and protection activities, tasks such as full geomorphology mapping and underwater area sweeping in turbid water pose serious challenges ...

Abstract: This study addresses a variant of the Vehicle Routing Problem (VRP) with customer priorities. In the variant, we assume the hard priority constraint where customers should be served in a ...