Yupu Lu

Ting Xu, Xu He, Yupu Lu et al.

This paper investigates the entropy dynamics of Chain-of-Thought (CoT) and uncovers a consistent two-phase structure: an Uncertainty Region of exploration transitioning sharply to a Confidence Region of convergence. We demonstrate that the Confidence Region possesses two critical properties: 1) High Reliability -- answers in the confidence region become highly accurate and stable, and 2) High Redundancy -- models generate unnecessary tokens long after reaching the correct answer. These properties unlock more efficient and reliable inference strategies: 1) Early Exit leverages reliability and redundancy to terminate computation safely when returns diminish, and 2)Test-Time Scaling uses the Confidence Region signal to prioritize converged trajectories. To operationalize these insights, we formulate Confidence Region detection as a sequential change-point detection problem, being the first to apply classical change-point methods to monitor CoT reasoning. Using the Cumulative Sum (CUSUM) algorithm, a statistically optimal change-point detector, we develop a training-free framework for real-time inference control. Experiments show our approach establishes a superior Pareto-frontier for early exit. CUSUM achieves 63.06% accuracy with 11.1% token reduction, outperforming DEER and Dynasor by 3.28% and 4.36% in accuracy respectively. For test-time scaling, CUSUM-weighted voting consistently outperforms self-consistency.

Yupu Lu, Shijie Lin, Guanqi Chen et al.

Deep learning models are able to approximate one specific dynamical system but struggle at learning generalisable dynamics, where dynamical systems obey the same laws of physics but contain different numbers of elements (e.g., double- and triple-pendulum systems). To relieve this issue, we proposed the Modular Lagrangian Network (ModLaNet), a structural neural network framework with modularity and physical inductive bias. This framework models the energy of each element using modularity and then construct the target dynamical system via Lagrangian mechanics. Modularity is beneficial for reusing trained networks and reducing the scale of networks and datasets. As a result, our framework can learn from the dynamics of simpler systems and extend to more complex ones, which is not feasible using other relevant physics-informed neural networks. We examine our framework for modelling double-pendulum or three-body systems with small training datasets, where our models achieve the best data efficiency and accuracy performance compared with counterparts. We also reorganise our models as extensions to model multi-pendulum and multi-body systems, demonstrating the intriguing reusable feature of our framework.

Yupu Lu, Shijie Lin, Hao Xu et al.

Finding a control Lyapunov function (CLF) in a dynamical system with a controller is an effective way to guarantee stability, which is a crucial issue in safety-concerned applications. Recently, deep learning models representing CLFs have been applied into a learner-verifier framework to identify satisfiable candidates. However, the learner treats Lyapunov conditions as complex constraints for optimisation, which is hard to achieve global convergence. It is also too complicated to implement these Lyapunov conditions for verification. To improve this framework, we treat Lyapunov conditions as inductive biases and design a neural CLF and a CLF-based controller guided by this knowledge. This design enables a stable optimisation process with limited constraints, and allows end-to-end learning of both the CLF and the controller. Our approach achieves a higher convergence rate and larger region of attraction (ROA) in learning the CLF compared to existing methods among abundant experiment cases. We also thoroughly reveal why the success rate decreases with previous methods during learning.

Sihan Chen, Hang Xu, Yupu Lu et al.

Shared autonomy (SA) enables robots to infer human intent and assist in its achievement. While most research focuses on improving intent inference, it overlooks whether humans can understand the robot's intent in return. Without such mutual understanding, collaboration becomes less effective, degrading user experience and task performance. To address this gap, previous studies have explicitly conveyed the robot intent through additional interfaces, which remain unintuitive and limited in expressiveness. Inspired by impedance control, we propose Impedance-Driven Anisotropic Guidance Field Enhanced Shared Autonomy (IAGF-SA), a novel paradigm that extends SA with an embodied, physically-grounded communication channel. This channel adaptively modulates the robot's dynamic response to human input, enabling intuitive, continuous, physically-grounded robot intent communication while naturally guiding human actions. User studies across three scenarios and two teleoperation interfaces indicate that IAGF-SA improves task performance, human-robot agreement, and subjective experience, thus demonstrating its effectiveness in enhancing human-robot communication and collaboration.

Yupu Lu, Yuxiang Ma, Jia Pan

This paper presents RichMap, a high-precision reachability map representation designed to balance efficiency and flexibility for versatile robot manipulation tasks. By refining the classic grid-based structure, we propose a streamlined approach that achieves performance close to compact map forms (e.g., RM4D) while maintaining structural flexibility. Our method utilizes theoretical capacity bounds on $\mathbb{S}^2$ (or $SO(3)$) to ensure rigorous coverage and employs an asynchronous pipeline for efficient construction. We validate the map against comprehensive metrics, pursuing high prediction accuracy ($>98\%$), low false positive rates ($1\sim2\%$), and fast large-batch query ($\sim$15 $μ$s/query). We extend the framework applications to quantify robot workspace similarity via maximum mean discrepancy (MMD) metrics and demonstrate energy-based guidance for diffusion policy transfer, achieving up to $26\%$ improvement for cross-embodiment scenarios in the block pushing experiment.

Sławomir Konrad Tadeja, Krzysztof Kutt, Yupu Lu et al.

In this paper, we present a Virtual Reality (VR) based environment where the engineer interacts with incoming data from a fleet of aeroengines. This data takes the form of 3D computer-aided design (CAD) engine models coupled with characteristic plots for the subsystems of each engine. Both the plots and models can be interacted with and manipulated using speech or gestural input. The characteristic data is ported to a knowledge-based system underpinned by a knowledge-graph storing complex domain knowledge. This permits the system to respond to queries about the current state and health of each aeroengine asset. Responses to these questions require some degree of analysis, which is handled by a semantic knowledge representation layer managing information on aeroengine subsystems. This paper represents a significant step forward for aeroengine analysis in a bespoke VR environment and brings us a step closer to a Jarvis-like system for aeroengine analytics.

Slawomir Konrad Tadeja, Wojciech Rydlewicz, Yupu Lu et al.

Photogrammetry is a science dealing with obtaining reliable information about physical objects using their imagery description. Recent advancements in the development of Virtual Reality (VR) can help to unlock the full potential offered by the digital 3D-reality models generated using the state-of-art photogrammetric technologies. These models are becoming a viable alternative for providing high-quality content for such immersive environment. Simultaneously, their analyses in VR could bring added-value to professionals working in various engineering and non-engineering settings and help in extracting useful information about physical objects. However, there is little research published to date on feasible interaction methods in the VR-based systems augmented with the 3D photogrammetric models, especially concerning gestural input interfaces. Consequently, this paper presents the PhotoTwinVR -- an immersive, gesture-controlled system for manipulation and inspection of 3D photogrammetric models of physical objects in VR. Our system allows the user to perform basic engineering operations on the model subjected to the off-line inspection process. An observational study with a group of three domain-expert participants was completed to verify its feasibility. The system was populated with a 3D photogrammetric model of an existing pipe-rack generated using a commercial software package. The participants were asked to carry out a survey measurement of the object using the measurement toolbox offered by PhotoTwinVR. The study revealed a potential of such immersive tool to be applied in practical real-words cases of off-line inspections of pipelines.