Users of feature-rich tools like Excel often miss more efficient workflows, repeating tedious steps and making avoidable errors. Current approaches to helping them require either manual prompting, which is effortful for users, or automated logging, which is limiting for developers. We present InvisibleMentor, a system inspired by over-the-shoulder learning: it observes what users do, then shows them how to do it better. To do this, InvisibleMentor analyzes screen recordings with a vision-language model to reconstruct actions and context, then uses a large language model to generate vision-grounded task reflection, structured suggestions grounded in observed behavior. In a user study, participants found InvisibleMentor’s suggestions more clear, more relevant, and more useful than those from a prompt-based assistant, demonstrating that AI can do more than automate away work—it can help users master it.

Although the latest artificial intelligence technologies can greatly improve work efficiency by automatically generating feasible solutions in the digital world (DW), they are incapable of discovering or creating new knowledge, i.e., lack of human intelligence or creativity. To break this limitation, this article describes and elaborates the masterplan of the digital intelligent world (DIW), wherein everyone has an intelligent agent (IA) for searching, exchanging, and processing information and knowledge autonomously. First, a data-information-knowledge-intelligence (DIKI) model is proposed to illustrate the challenges of creating intelligence from raw data, and of realizing the DIW from the DW. Specifically, the DIW adopts knowledge-driven approaches and could achieve huge productivity enhancement through cross-domain innovations and deep intelligentization with broader creativity. Second, at the individual level, a knowledge processing architecture of IA is defined to support knowledge-centric operations and services. Third, at the system level, a framework of knowledge market (KM) is established for fair, effective, and autonomous collaborations among massive IAs. Inspired by basic laws in statistical thermodynamics, information sciences, and economics, three fundamental principles are developed and discussed for guaranteeing a prosperous KM and the sustainable DIW.

Programming assistants have reshaped the experience of programming into one where programmers spend less time writing and more time critically examining code. In this paper, we explore how programming assistants can be extended to accelerate the inspection of generated code. We introduce an extension to the programming assistant called Ivie, or instantly visible in-situ explanations. When using Ivie, a programmer’s generated code is instantly accompanied by explanations positioned just adjacent to the code. Our design was optimized for low-cost invocation and dismissal. Explanations are compact and informative. They describe meaningful expressions, from individual variables to entire blocks of code. We present an implementation of Ivie that forks VS Code, applying a modern LLM for timely segmentation and explanation of generated code. In a lab study, we compared Ivie to a contemporary baseline tool for code understanding. Ivie improved understanding of generated code, and was received by programmers as a highly useful, low distraction complement to the programming assistant.

The radio stripe (RS) is an emerging architecture for cell-free massive multiple-input-multiple-output (CFm-MIMO) communication systems, where access points (APs) are sequentially connected via one optical-fiber link. The traditional precoded signal vector quantization approach is designed for star-topology MIMO architecture with a large demand for fronthaul capacity thereby it is not suitable for RS architecture. To address this challenge, a joint precoding matrix optimization and pruning approach is proposed to reduce the fronthaul traffic in the RS architecture. Central to our strategy is optimizing and pruning the precoding coefficient matrix (PCM), which is decomposed from the stationary point of the weighted sum-rate maximization problem. An iterative algorithm with closed-form iteration expression is developed for the joint optimization and pruning of PCMs. Note that the size of PCM only depends on the number of data streams rather than the antenna number. In contrast to transmitting different precoding matrices to different APs, the required PCMs for different APs are proved to be the same, which illustrates that only one PCM needs to be transmitted in the RS. Simulation results demonstrate that the proposed approach can reduce the fronthaul link traffic by 739% compared to the weighted minimum mean square error (WMMSE) precoding algorithm. When each signal symbol is quantized with 10 bits, the proposed approach can improve the sum-rate by 284% and 544% compared to the zero-forcing and maximum ratio transmission algorithms, respectively.

Mobile computing devices are supporting higher information transmission and processing rates. However, the massive amount of information transmitted and processed also means that a vast amount of heat is produced. Due to the constraints of safety and portability, the problem of overheating is becoming a main challenge for mobile computing devices to achieve their ideal performance. Compared to studies on improving the quality of wireless communications and mobile computing, little attention has been paid to the temperature and its direct impact on the computing performance of the device. In this paper, we propose the heat transfer model of the processor of a mobile computing device based on thermodynamics. Considering Landauer’s principle, the maximum computing rate of the processor is derived. Moreover, the performance of devices adopting multi-processor is evaluated, where each processor works periodically to complete the computing task. The conditions for stable computing and the thermodynamic advantage of multi-processor are given. Based on the evaluation, an optimization method is proposed to lower the average temperature of the processor, and simulation results show that the maximum computing rate can be improved up to 20.9%.

One of the most prominent physical aspects of mobile communication systems is that they are inherently non-equilibrium systems. Traditional researches on the energetic costs of communication systems pay little attention to the relationship between the energy and the information transmitted and processed by the system. On the other hand, recent breakthroughs in nonequilibrium thermodynamics have led to a deeper understanding of the thermodynamics of information. To investigate the energetic costs of a mobile communication system at a fundamental level, in this paper, an entropy-based energy dissipation model based on nonequilibrium thermodynamics is first proposed for mobile communication systems. The energy dissipation model relates the energy and information through the common concept “entropy” from thermodynamics and information theory. Moreover, the theoretical minimal energy dissipation limits are derived for typical modulations in mobile communication systems. Simulation results show that the practical energy dissipation of information processing and information transmission is three and seven orders of magnitude away from the theoretical minimal energy dissipation limits in mobile communication systems, respectively. The energy dissipation model and results derived in the paper provide guidelines on how to design future mobile communication systems to minimize the thermodynamic costs.

Programmers often rely on online resources—such as code examples, documentation, blogs, and Q&A forums—to compare similar libraries and select the one most suitable for their own tasks and contexts. However, this comparison task is often done in an ad-hoc manner, which may result in suboptimal choices. Inspired by Analogical Learning and Variation Theory, we hypothesize that rendering many concept-annotated code examples from different libraries side-by-side can help programmers (1) develop a more comprehensive understanding of the libraries’ similarities and distinctions and (2) make more robust, appropriate library selections. We designed a novel interactive interface, ParaLib, and used it as a technical probe to explore to what extent many side-by-side concepted-annotated examples can facilitate the library comparison and selection process. A within-subjects user study with 20 programmers shows that, when using ParaLib, participants made more consistent, suitable library selections and provided more comprehensive summaries of libraries’ similarities and differences.

Many programmers want to use deep learning due to its superior accuracy in many challenging domains. Yet our formative study with ten programmers indicated that, when constructing their own deep neural networks (DNNs), they often had a difficult time choosing appropriate model structures and hyperparameter values. This paper presents ExampleNet—a novel interactive visualization system for exploring common and uncommon design choices in a large collection of open-source DNN projects. ExampleNet provides a holistic view of the distribution over model structures and hyperparameter settings in the corpus of DNNs, so users can easily filter the corpus down to projects tackling similar tasks and compare design choices made by others. We evaluated ExampleNet in a within-subjects study with sixteen participants. Compared with the control condition (i.e., online search), participants using ExampleNet were able to inspect more online examples, make more data-driven design decisions, and make fewer design mistakes.

Tagged magnetic resonance (MR) imaging has been shown to be a useful means for non-invasive measurement of the deformation of the left ventricle (LV) during the cardiac cycle. One can estimate a dense displacement field based on the tag line movements and use it to compute myocardial measures such as strain. Existing methods require that the boundaries of the LV be known before the displacement field is estimated, which results in a time-consuming process since user intervention is needed for the estimation of the boundaries. In this paper, the authors propose a method that jointly estimates the LV boundaries and displacement field without user-intervention. They model the displacement field as a compound Gauss-Markov random field (CGMRF) which is parameterized by two closed and smooth contours. A partial-optimal solution is sought by iteratively updating the displacement field, the contours and the parameters. Experimental results on both simulated vector image and in vivo human data show that the authors' method is capable of automatically tracking the contours and reconstructing displacement field with a decent accuracy.