Zhaoxiang Li

Bingcheng Hu, Lixiang Jin, Zhaoxiang Li

In this paper, we propose a novel Physics-Informed Neural Network (PINN) framework based on the Cordès condition for solving both linear and fully nonlinear partial differential equations (PDEs) in non-divergence form, together with their applications. By incorporating the operator structure into the loss function, the proposed method improves the conditioning of the associated optimization problem, thereby enhancing training stability and solution accuracy. The framework is further extended to include Hamilton-Jacobi-Bellman and Monge-Ampère equations, with applications to optimal transport. Numerical experiments demonstrate the effectiveness and robustness of the method, as well as its capability to address high-dimensional problems, highlighting the promise of learning-based approaches for tackling challenging PDEs. Owing to its generality and simplicity, the proposed method is expected to be of broad interest to the scientific and engineering communities.

Qingyu Fan, Zhaoxiang Li, Yi Lu et al.

Bimanual manipulation in cluttered scenes requires policies that remain stable under occlusions, viewpoint and scene variations. Existing vision-language-action models often fail to generalize because (i) multi-view features are fused via view-agnostic token concatenation, yielding weak 3D-consistent spatial understanding, and (ii) language is injected as global conditioning, resulting in coarse instruction grounding. In this paper, we introduce PEAfowl, a perception-enhanced multi-view VLA policy for bimanual manipulation. For spatial reasoning, PEAfowl predicts per-token depth distributions, performs differentiable 3D lifting, and aggregates local cross-view neighbors to form geometrically grounded, cross-view consistent representations. For instruction grounding, we propose to replace global conditioning with a Perceiver-style text-aware readout over frozen CLIP visual features, enabling iterative evidence accumulation. To overcome noisy and incomplete commodity depth without adding inference overhead, we apply training-only depth distillation from a pretrained depth teacher to supervise the depth-distribution head, providing perception front-end with geometry-aware priors. On RoboTwin 2.0 under domain-randomized setting, PEAfowl improves the strongest baseline by 23.0 pp in success rate, and real-robot experiments further demonstrate reliable sim-to-real transfer and consistent improvements from depth distillation. Project website: https://peafowlvla.github.io/.