**SingularTrajectory: Universal Trajectory Predictor Using Diffusion Model** Inhwan Bae, Young-Jae Park, and Hae-Gon Jeon from the AI Graduate School at GIST, South Korea, propose SingularTrajectory, a diffusion-based universal trajectory prediction framework. The paper addresses five types of trajectory prediction tasks: deterministic, stochastic, domain adaptation, momentary observation, and few-shot. Despite the common input and output formats, specialized architectures are often required for each task, leading to sub-optimal performance. SingularTrajectory aims to unify these tasks by creating a Singular space that projects various motion patterns into a single embedding space. An adaptive anchor in this space corrects incorrect anchors based on a traversability map, enhancing the accuracy of predicted paths. A diffusion-based predictor further refines these paths using a cascaded denoising process. Extensive experiments on five public benchmarks demonstrate that SingularTrajectory significantly outperforms existing models, highlighting its effectiveness in estimating general human movement dynamics. The code is publicly available at <https://github.com/inhwanbae/SingularTrajectory>.**SingularTrajectory: Universal Trajectory Predictor Using Diffusion Model** Inhwan Bae, Young-Jae Park, and Hae-Gon Jeon from the AI Graduate School at GIST, South Korea, propose SingularTrajectory, a diffusion-based universal trajectory prediction framework. The paper addresses five types of trajectory prediction tasks: deterministic, stochastic, domain adaptation, momentary observation, and few-shot. Despite the common input and output formats, specialized architectures are often required for each task, leading to sub-optimal performance. SingularTrajectory aims to unify these tasks by creating a Singular space that projects various motion patterns into a single embedding space. An adaptive anchor in this space corrects incorrect anchors based on a traversability map, enhancing the accuracy of predicted paths. A diffusion-based predictor further refines these paths using a cascaded denoising process. Extensive experiments on five public benchmarks demonstrate that SingularTrajectory significantly outperforms existing models, highlighting its effectiveness in estimating general human movement dynamics. The code is publicly available at <https://github.com/inhwanbae/SingularTrajectory>.