Adaptive Training for Nautical Rules of the Road
研究通过对比自适应与非自适应船舶驾驶模拟训练，发现自适应训练能更有效提升学生对航海规则的知识掌握和应用能力。
0 citationsRead paper
Selected work
Representative Papers
A Runtime Decentralized Attestation and Coordinated Repair Framework for Securing Automotive ECUs
This work addresses the vulnerability of vehicle electronic control units (ECUs) to malware attacks and the challenge of achieving both efficient runtime detection and secure recovery with existing approaches. To this end, the paper proposes DACER, a novel framework that co-designs local firmware rollback with global ECU reboot mechanisms, leveraging the hierarchical nature of in-vehicle computing architectures to enable runtime recovery without single points of failure while meeting real-time constraints. Built upon ARM TrustZone and a secure flash memory controller, DACER supports per-ECU self-authentication, self-recovery, and low-overhead distributed coordination. Experimental evaluation on real hardware demonstrates that the framework efficiently performs whole-vehicle state verification and firmware restoration with minimal runtime overhead.
0 citationsRead paper
PromptPath: Prompt-Adaptive Computational Pathways for In-Context Learning
This work addresses the limitations of existing in-context learning approaches, which treat prompts merely as semantic cues and thus fail to enable task-adaptive dynamic computation, resulting in shallow and uninterpretable reasoning. To overcome this, the authors propose PromptPath, a novel framework that directly integrates prompt information into the model’s inference architecture. PromptPath employs a prompt-conditioned routing mechanism to dynamically activate and compose lightweight low-rank expert modules, thereby constructing task-specific computational pathways. This approach achieves dynamic adaptation at the computational level, significantly outperforming current methods on both 3D point cloud and 2D visual recognition benchmarks while demonstrating strong cross-domain and cross-task generalization capabilities.
0 citationsRead paper
Receding-Horizon Next-Best-View Planner for Autonomous Leaf Surface Reconstruction
This work addresses the challenge of high-fidelity reconstruction of plant leaf surfaces in resource-constrained field environments by proposing a receding-horizon next-best-view (NBV) planning approach. The method introduces a centroid-based information gain function to quantify view utility and integrates multi-step lookahead reasoning to optimize the robot’s observation trajectory, effectively balancing computational efficiency with reconstruction quality while mitigating inter-leaf occlusions. Experimental results on strawberry plants across multiple growth stages demonstrate that the proposed approach significantly reduces surface reconstruction error and enhances geometric fidelity, achieving up to a 10% improvement in reconstruction accuracy over baseline methods.
0 citationsRead paper
grapheme-kit: Grapheme-Level Metrics and Text Processing for Multilingual NLP
This work addresses the limitations of existing Unicode code point–based text evaluation methods, which often fail to accurately measure character-level errors in complex writing systems where a single grapheme frequently comprises multiple code points. To overcome this, the authors introduce grapheme-kit, an open-source Python library that, for the first time, extends widely used NLP evaluation metrics—such as edit distance and similarity—to the grapheme cluster level. By adhering to Unicode standards for grapheme cluster identification, composition, and decomposition, the proposed approach significantly improves evaluation accuracy for tasks like OCR on scripts with complex orthographies, including Tamil and Sinhala. This advancement provides a precise, grapheme-aware toolkit for text processing in low-resource languages.
0 citationsRead paper
Recent publications
Latest Papers
Adaptive Training for Nautical Rules of the Road
研究通过对比自适应与非自适应船舶驾驶模拟训练，发现自适应训练能更有效提升学生对航海规则的知识掌握和应用能力。
0 citationsRead paper
A Runtime Decentralized Attestation and Coordinated Repair Framework for Securing Automotive ECUs
This work addresses the vulnerability of vehicle electronic control units (ECUs) to malware attacks and the challenge of achieving both efficient runtime detection and secure recovery with existing approaches. To this end, the paper proposes DACER, a novel framework that co-designs local firmware rollback with global ECU reboot mechanisms, leveraging the hierarchical nature of in-vehicle computing architectures to enable runtime recovery without single points of failure while meeting real-time constraints. Built upon ARM TrustZone and a secure flash memory controller, DACER supports per-ECU self-authentication, self-recovery, and low-overhead distributed coordination. Experimental evaluation on real hardware demonstrates that the framework efficiently performs whole-vehicle state verification and firmware restoration with minimal runtime overhead.
0 citationsRead paper
PromptPath: Prompt-Adaptive Computational Pathways for In-Context Learning
This work addresses the limitations of existing in-context learning approaches, which treat prompts merely as semantic cues and thus fail to enable task-adaptive dynamic computation, resulting in shallow and uninterpretable reasoning. To overcome this, the authors propose PromptPath, a novel framework that directly integrates prompt information into the model’s inference architecture. PromptPath employs a prompt-conditioned routing mechanism to dynamically activate and compose lightweight low-rank expert modules, thereby constructing task-specific computational pathways. This approach achieves dynamic adaptation at the computational level, significantly outperforming current methods on both 3D point cloud and 2D visual recognition benchmarks while demonstrating strong cross-domain and cross-task generalization capabilities.
0 citationsRead paper
Receding-Horizon Next-Best-View Planner for Autonomous Leaf Surface Reconstruction
This work addresses the challenge of high-fidelity reconstruction of plant leaf surfaces in resource-constrained field environments by proposing a receding-horizon next-best-view (NBV) planning approach. The method introduces a centroid-based information gain function to quantify view utility and integrates multi-step lookahead reasoning to optimize the robot’s observation trajectory, effectively balancing computational efficiency with reconstruction quality while mitigating inter-leaf occlusions. Experimental results on strawberry plants across multiple growth stages demonstrate that the proposed approach significantly reduces surface reconstruction error and enhances geometric fidelity, achieving up to a 10% improvement in reconstruction accuracy over baseline methods.
0 citationsRead paper
grapheme-kit: Grapheme-Level Metrics and Text Processing for Multilingual NLP
This work addresses the limitations of existing Unicode code point–based text evaluation methods, which often fail to accurately measure character-level errors in complex writing systems where a single grapheme frequently comprises multiple code points. To overcome this, the authors introduce grapheme-kit, an open-source Python library that, for the first time, extends widely used NLP evaluation metrics—such as edit distance and similarity—to the grapheme cluster level. By adhering to Unicode standards for grapheme cluster identification, composition, and decomposition, the proposed approach significantly improves evaluation accuracy for tasks like OCR on scripts with complex orthographies, including Tamil and Sinhala. This advancement provides a precise, grapheme-aware toolkit for text processing in low-resource languages.
0 citationsRead paper