Previouswork onneural 3Dreconstruction demonstrated benefits, butalso limitations, ofpoint cloud, voxel, surface mesh, and implicit function representations. Unlike existing volumetric approaches,DEFTET optimizes for both vertex placement and occupancy, and is differentiable with respect to standard 3D reconstruction lossfunctions.

We sequentially add regularizations from left to right. We report 3D IOU (higher is better). We report Chamfer Distance (lower is better). This loss is applied only when 3D ground-truth is available. We find that all of these regularizations are necessary to obtain non-degenerated and smooth predictions, as shown next.

Modular robotics research has long been preoccupied with perfecting the modules themselves -- their actuation methods, connectors, controls, communication, and fabrication. This inward focus results, in part, from the complexity of the task and largely confines modular robots to sterile laboratory settings. The latest generation of truss modular robots, such as the Variable Topology Truss and the Truss Link, have begun to focus outward and reveal a key insight: the environment is not just a backdrop; it is a tool. In this work, we shift the paradigm from building better robots to building better robot environment interactions for modular truss robots. We study how modular robots can effectively exploit their surroundings to achieve faster locomotion, adaptive self-reconfiguration, and complex three-dimensional assembly from simple two-dimensional robot assemblies. By using environment features -- ledges, gaps, and slopes -- we show how the environment can extend the robots' capabilities. Nature has long mastered this principle: organisms not only adapt, but exploit their environments to their advantage. Robots must learn to do the same. This study is a step towards modular robotic systems that transcend their limitations by exploiting environmental features.

We use machine learning to search for examples of Z/2 eigenfunctions on the 2-sphere. For this we created a multivalued version of a feedforward deep neural network, and we implemented it using the JAX library. We found Z/2 eigenfunctions for three cases: In the first two cases we fixed the branch points at the vertices of a tetrahedron and at a cube respectively. In a third case, we allowed the AI to move the branch points around and, in the end, it positioned the branch points at the vertices of a squashed tetrahedron.

In this paper, we investigate code-integrated reasoning, where models generate code when necessary and integrate feedback by executing it through a code interpreter. To acquire this capability, models must learn when and how to use external code tools effectively, which is supported by tool-augmented reinforcement learning (RL) through interactive learning. Despite its benefits, tool-augmented RL can still suffer from potential instability in the learning dynamics. In light of this challenge, we present a systematic approach to improving the training effectiveness and stability of tool-augmented RL for code-integrated reasoning. Specifically, we develop enhanced training strategies that balance exploration and stability, progressively building tool-use capabilities while improving reasoning performance. Through extensive experiments on five mainstream mathematical reasoning benchmarks, our model demonstrates significant performance improvements over multiple competitive baselines. Furthermore, we conduct an in-depth analysis of the mechanism and effect of code-integrated reasoning, revealing several key insights, such as the extension of model's capability boundaries and the simultaneous improvement of reasoning efficiency through code integration. All data and code for reproducing this work are available at: https://github.com/RUCAIBox/CIR.

In recent years, soft robotics simulators have evolved to offer various functionalities, including the simulation of different material types (e.g., elastic, hyper-elastic) and actuation methods (e.g., pneumatic, cable-driven, servomotor). These simulators also provide tools for various tasks, such as calibration, design, and control. However, efficiently and accurately computing derivatives within these simulators remains a challenge, particularly in the presence of physical contact interactions. Incorporating these derivatives can, for instance, significantly improve the convergence speed of control methods like reinforcement learning and trajectory optimization, enable gradient-based techniques for design, or facilitate end-to-end machine-learning approaches for model reduction. This paper addresses these challenges by introducing a unified method for computing the derivatives of mechanical equations within the finite element method framework, including contact interactions modeled as a nonlinear complementarity problem. The proposed approach handles both collision and friction phases, accounts for their nonsmooth dynamics, and leverages the sparsity introduced by mesh-based models. Its effectiveness is demonstrated through several examples of controlling and calibrating soft systems.