arXiv subjects

Yuchen Xi

Publications and source records attributed to Yuchen Xi.

6 recordsLinked to original sources

arXiv2026

TRACE-GS: On-Policy Trajectory Distillation with Privileged Geometric Conditioning for Sparse-View 3DGS Restoration

We present TRACE-GS, an on-policy trajectory distillation framework that leverages privileged geometric conditioning at training time, thereby adapting a diffusion prior to sparse-view 3D Gaussian Splatting (3DGS) restoration. Rather than pursuing increasingly sophisticated restoration architectures, we identify a more fundamental limitation shared by existing diffusion-based approaches: supervision at independently noised states does not cover those reached during inference. In sparse-view 3DGS, under-constrained geometry biases denoising from the outset, and the resulting deviations compound along the rollout. TRACE-GS instead performs on-policy trajectory distillation: a teacher conditioned on richer geometry from additional training views supplies targets along the sparse-view student's own rollout, aligning denoising directions and cross-view responses at each visited state. This training-only geometry places TRACE-GS in the learning using privileged information (LUPI) setting. At deployment, only the sparse-view student is retained, and its restored renderings serve as pseudo-observations for 3DGS refinement. To the best of our knowledge, TRACE-GS is the first to derive on-policy supervision from privileged geometry for sparse-view 3DGS restoration, achieving consistent gains and strong generalization across datasets and sparse-view settings.

cs.CV↗

arXiv2026

The physics of cr\^epes: Elasto-gravity control of soft folding

Like a cr\^epe resting on a plate, a thin elastic sheet can fold smoothly under its own weight, forming reversible shapes without creases or imposed hinges. Such soft folds arise from a balance between elastic bending and gravity, yet their stability, packing limits, and dynamics remain poorly understood. Here we show that these behaviors are governed by a single physical length scale, the elasto-gravity length $\ell_{eg}$. Using experiments and heavy-elastica theory, we demonstrate that $\ell_{eg}$ sets the characteristic fold geometry, determines when a fold becomes unstable and unfolds, and limits how many reversible folds can be stacked in rectangular and circular sheets. In particular, when lengths are rescaled by $\ell_{eg}$, fold shapes and stability thresholds collapse across materials and thicknesses. We further show that unfolding follows a universal speed scaling $v \sim \sqrt{g\,\ell_{eg}}$, revealing a gravity-controlled time scale for the release of stored bending energy. Together, these results establish a unified physical framework for reversible folding, compact storage, and gravity-assisted deployment of thin elastic sheets.

cond-mat.soft↗

arXiv2025

Building granular structures with elasto-active systems

Natural active systems routinely reshape and reorganize their environments through sustained local interactions. Examples of decentralized collective construction are common in nature, e.g., many insects achieve large-scale constructions through indirect communication. While synthetic realizations of self-organization exist, they typically rely on rigid agents that require some kind of sensors and direct programming to achieve their function. Understanding how soft, deformable active matter navigates and remodels crowded landscapes remains an open challenge. Here we show that connecting rigid microbots to elastic beams yields elasto-active structures that can restructure and adapt to heterogeneous surroundings. We investigate the dynamics of these agents in environments with varying granular densities, rationalizing how they can aggregate or carve the medium through gentle interactions. At low density, the system compacts dispersed obstacles into clusters, a process modeled by a modified Smoluchowski coagulation theory. At high density, our agents carve voids whose size is predicted by a force-limited argument. These results establish a framework for understanding how activity, elasticity, and deformability can influence active navigation and environmental reconfiguration in granular media.

cond-mat.soft↗

arXiv2025

MMRHP: A Miniature Mixed-Reality HIL Platform for Auditable Closed-Loop Evaluation

Validation of autonomous driving systems requires a trade-off between test fidelity, cost, and scalability. While miniaturized hardware-in-the-loop (HIL) platforms have emerged as a promising solution, a systematic framework supporting rigorous quantitative analysis is generally lacking, limiting their value as scientific evaluation tools. To address this challenge, we propose MMRHP, a miniature mixed-reality HIL platform that elevates miniaturized testing from functional demonstration to rigorous, reproducible quantitative analysis. The core contributions are threefold. First, we propose a systematic three-phase testing process oriented toward the Safety of the Intended Functionality(SOTIF)standard, providing actionable guidance for identifying the performance limits and triggering conditions of otherwise correctly functioning systems. Second, we design and implement a HIL platform centered around a unified spatiotemporal measurement core to support this process, ensuring consistent and traceable quantification of physical motion and system timing. Finally, we demonstrate the effectiveness of this solution through comprehensive experiments. The platform itself was first validated, achieving a spatial accuracy of 10.27 mm RMSE and a stable closed-loop latency baseline of approximately 45 ms. Subsequently, an in-depth Autoware case study leveraged this validated platform to quantify its performance baseline and identify a critical performance cliff at an injected latency of 40 ms. This work shows that a structured process, combined with a platform offering a unified spatio-temporal benchmark, enables reproducible, interpretable, and quantitative closed-loop evaluation of autonomous driving systems.

cs.RO↗

arXiv2025

ROAD: Responsibility-Oriented Reward Design for Reinforcement Learning in Autonomous Driving

Reinforcement learning (RL) in autonomous driving employs a trial-and-error mechanism, enhancing robustness in unpredictable environments. However, crafting effective reward functions remains challenging, as conventional approaches rely heavily on manual design and demonstrate limited efficacy in complex scenarios. To address this issue, this study introduces a responsibility-oriented reward function that explicitly incorporates traffic regulations into the RL framework. Specifically, we introduced a Traffic Regulation Knowledge Graph and leveraged Vision-Language Models alongside Retrieval-Augmented Generation techniques to automate reward assignment. This integration guides agents to adhere strictly to traffic laws, thus minimizing rule violations and optimizing decision-making performance in diverse driving conditions. Experimental validations demonstrate that the proposed methodology significantly improves the accuracy of assigning accident responsibilities and effectively reduces the agent's liability in traffic incidents.

cs.LG↗

arXiv2024

Emergent intelligence of buckling-driven elasto-active structures

Active systems of self-propelled agents, e.g., birds, fish, and bacteria, can organize their collective motion into myriad autonomous behaviors. Ubiquitous in nature and across length scales, such phenomena are also amenable to artificial settings, e.g., where brainless self-propelled robots orchestrate their movements into spatio-temportal patterns via the application of external cues or when confined within flexible boundaries. Very much like their natural counterparts, these approaches typically require many units to initiate collective motion such that controlling the ensuing dynamics is challenging. Here, we demonstrate a novel yet simple mechanism that leverages nonlinear elasticity to tame near-diffusive motile particles in forming structures capable of directed motion and other emergent intelligent behaviors. Our elasto-active system comprises two centimeter-sized self-propelled microbots connected with elastic beams. These microbots exert forces that suffice to buckle the beam and set the structure in motion. We first rationalize the physics of the interaction between the beam and the microbots. Then we use reduced order models to predict the interactions of our elasto-active structure with boundaries, e.g., walls and constrictions, and demonstrate how they can exhibit intelligent behaviors such as maze navigation. The findings are relevant to designing intelligent materials or soft robots capable of autonomous space exploration, adaptation, and interaction with the surrounding environment.

cond-mat.soft↗