A5031284816- Software Engineering Research
- Open Source Software Innovations
- Reinforcement Learning in Robotics
- Software Reliability and Analysis Research
- Software Engineering Techniques and Practices
- Robot Manipulation and Learning
- Modular Robots and Swarm Intelligence
- Multimodal Machine Learning Applications
- Brain Metastases and Treatment
- Human Pose and Action Recognition
- Technology and Security Systems
- IoT and Edge/Fog Computing
- Advanced Neural Network Applications
- Natural Language Processing Techniques
- Lung Cancer Treatments and Mutations
- Geographic Information Systems Studies
- Software System Performance and Reliability
- Cryptography and Data Security
- Ideological and Political Education
- Technology Adoption and User Behaviour
- Customer Service Quality and Loyalty
- Business Process Modeling and Analysis
- Big Data and Business Intelligence
- Service-Oriented Architecture and Web Services
- Cloud Data Security Solutions
Beijing Chao-Yang Hospital
2024
Capital Medical University
2024
China Nonferrous Metal Mining (China)
2024
PLA Academy of Military Science
2023
Jiamusi University
2023
Sun Yat-sen University
2005-2023
University of Groningen
2020-2022
ZTE (China)
2022
China Mobile (China)
2022
State Key Laboratory of Mobile Networks and Mobile Multimedia Technology
2022
Large language models (LLMs) have demonstrated exciting progress in acquiring diverse new capabilities through in-context learning, ranging from logical reasoning to code-writing. Robotics researchers also explored using LLMs advance the of robotic control. However, since low-level robot actions are hardware-dependent and underrepresented LLM training corpora, existing efforts applying robotics largely treated as semantic planners or relied on human-engineered control primitives interface...
Large, high-capacity models trained on diverse datasets have shown remarkable successes efficiently tackling downstream applications. In domains from NLP to Computer Vision, this has led a consolidation of pretrained models, with general backbones serving as starting point for many Can such happen in robotics? Conventionally, robotic learning methods train separate model every application, robot, and even environment. we instead generalist X-robot policy that can be adapted new robots,...
Safety is an essential component for deploying reinforcement learning (RL) algorithms in real-world scenarios, and critical during the process itself. A natural first approach toward safe RL to manually specify constraints on policy's behavior. However, just as has enabled progress large-scale development of AI systems, safety specifications may also be necessary ensure messy open-world environments where manual cannot scale. Akin how humans learn incrementally starting child-safe...
Brain metastases from malignant melanoma carry a poor prognosis. Novel systemic agents have improved overall survival (OS), but the value of whole-brain radiotherapy (WBRT) and stereotactic radiosurgery (SRS) remains uncertain. The melanoma-specific graded prognostic assessment (msGPA) provides useful information, relevance to modern-day population has not been validated. Since 2011, 53 patients received treatment for brain at Rosemere Cancer Centre medical oncology clinic. Data were...
Conceptual modeling, which includes both control-flow and data-flow has posed great challenges for conventional business process management systems. To support systematic modeling analysis, in this paper, we propose a novel effective Petri net-based approach. We first introduce new type of net, called WFIO-net, the firing rule is formally defined by extending classical net with read write semantics, to model information. Then, discuss about three possible errors over i.e., missing, redundant...
Large language models (LLMs) have demonstrated the potential to perform high-level planning. Yet, it remains a challenge for LLMs comprehend low-level commands, such as joint angle targets or motor torques. This paper proposes an approach use foot contact patterns interface that bridges human commands in natural and locomotion controller outputs these commands. results interactive system quadrupedal robots allows users craft diverse behaviors flexibly. We contribute LLM prompt design, reward...
Recent advancements in large multimodal models have led to the emergence of remarkable generalist capabilities digital domains, yet their translation physical agents such as robots remains a significant challenge. This report introduces new family AI purposefully designed for robotics and built upon foundation Gemini 2.0. We present Robotics, an advanced Vision-Language-Action (VLA) model capable directly controlling robots. Robotics executes smooth reactive movements tackle wide range...
Abstract In recent years, the evolution of software ecosystems and detection technical debt received significant attention by researchers from both industry academia. While a few studies that analyze various aspects already exist, to best our knowledge, there is no large‐scale study focuses on remediation over time in Python projects—that is, one most popular programming languages at moment. this paper, we 44 open‐source projects belonging Apache Software Foundation. We focus type amount...
Technical Debt (TD) can be paid back either by those that incurred it or others. We call the former self-fixed TD, and is particularly effective, as developers are experts in their own code best-suited to fix corresponding TD issues. To what extent self-fixed, which types of more likely remediation time shorter than non-self-fixed TD? This paper attempts answer these questions. It reports on an empirical study analyzes issues five (i.e., Code, Defect, Design, Documentation Test), captured...
Evaluation of the Customer Perceived Service Quality in a quantifying is most important part service management literature. Over past decade, SERVQUAL instrument has emerged as perhaps popular standardized questionnaire to measure quality. This paper starts with elaborating great significance carry out this Chinese educational institutions. Based on brief introduction and related review instrument, criteria it are re-designed order be suitable evaluate quality higher education. Moreover,...
Collaboration requires agents to align their goals on the fly. Underlying human ability with other is predict intentions of others and actively update own plans. We propose hierarchical predictive planning (HPP), a model-based reinforcement learning method for decentralized multiagent rendezvous. Starting pretrained, single-agent point navigation policies using noisy, high-dimensional sensor inputs like lidar, we first learn via self-supervision motion predictions all team. Next, HPP uses...
The impact of Technical Debt (TD) on software maintenance and evolution is great concern, but recent evidence shows that a considerable amount TD fixed by the same developers who introduced it; this termed self-fixed TD. This characteristic management can potentially team dynamics practices in managing However, initial based low-level source code analysis; casts some doubt whether practitioners repay their own debt intentionally under what circumstances. To address gap, we conducted an...
Most of the existing ways for strengthening users’ confidence in uploading private data to a cloud platform stress too much on security. On contrary, computing must be open and distributed. Moreover, it should provide highly robust, secure, quality services users. Therefore, establishing balance between performance security platforms is necessary. In this study, hybrid privacy protection solution based key policy-attribute-based encryption (ABE) cipher policy-ABE proposed. solution,...
Context: Although Technical Debt (TD) has increasingly gained attention in recent years, most studies exploring TD are based on a single source (e.g., code, code comments or issue trackers).Objective: Investigating information combined from different sources may yield insight that is more than the sum of its parts.In particular, we argue how items managed both trackers and software repositories (including commit messages) can shed some light what happens between commits incur those pay it...
Tool use is a hallmark of advanced intelligence, exemplified in both animal behavior and robotic capabilities. This paper investigates the feasibility imbuing robots with ability to creatively tools tasks that involve implicit physical constraints long-term planning. Leveraging Large Language Models (LLMs), we develop RoboTool, system accepts natural language instructions outputs executable code for controlling simulated real-world environments. RoboTool incorporates four pivotal components:...
After the SMURF method was proposed in 2006, many researchers have found inadequate of method; they new to improve this method. However, all work been done focus on is less concern mathematical theory. In paper, we analyze limitation view mathematics; some results received by analysis, way calculate parameter P <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i</sub> <sup xmlns:xlink="http://www.w3.org/1999/xlink">avg</sup> and improper value...
Technical debt (TD) reflects issues that may negatively affect software maintenance and evolution. There is currently little evidence on how the different types of TD co-occur; for example, code smells design same part system. This paper investigates co-occur, as well time period co-occurrence. To end, we analyzed co-occurring associations between five TD, captured in 42 SonarQube rules, 3862 files 20 Python projects from Apache Software Foundation. We found this phenomenon dominant,...
With the continuous development of IoT (internet things), era Internet everything is coming. In context everything, AI (artificial intelligence) technology plays a vital role in all fields. face massive data and equipment efficient performance requirements, new requirements are put forward for AI. Based on introduction standard, this paper illustrates symbiotic relationship between through application cases Furthermore, possible future research directions objectives field discussed, two will...