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 ...
A closed-loop framework for generating physically plausible and aesthetically coherent 3D indoor scenes through multi-turn iterative refinement. The system combines Vision-Language Model (VLM) ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results