Contemporary immersive theatre often limits audience agency to passive observation or spatial navigation, leaving social co-presence underdesigned. We present The Banquet, a co-located participatory sound theatre that turns culturally grounded ritual gestures—raising and clinking cups—into an interaction grammar for multi-user storytelling. Inspired by Dream of the Red Chamber, the system uses Interactive Audio Augmented Reality (IAAR) to embed narrative logic in embodied action, supported by high-precision tracking and a distributed Wi-Fi audio broadcast architecture. Layered spatial audio and subtle acoustic cues enable strangers to coordinate narrative progression without screens. We discuss how formalizing social rituals as tangible interfaces can support embodied negotiation and collective meaning-making in shared space.

Virtual Reality (VR) has been widely adopted in domains such as gaming, education, and healthcare, where 3D props play a central role in enabling immersive interaction. With the advancement of generative AI, 3D props can now be created rapidly; however, little research has explored how gestures and speech can be integrated to support prop generation. To address this gap, we introduce GestuProp, a VR prop generation system driven by co-speech gestures. Building on a formative study with 30 participants, we proposed a gesture design space and developed the VR system GestuProp. We then conducted a user study with 14 participants, which showed that GestuProp demonstrates good usability and favorable user experiences, while also revealing how object categories influence gesture use and interaction. These findings highlight the potential of gesture–speech synergy to advance prop generation in VR.

In this work, we present Patch-Adapter, an effective framework for high-resolution text-guided image inpainting. Unlike existing methods limited to lower resolutions, our approach achieves 4K+ resolution while maintaining precise content consistency and prompt alignment, two critical challenges in image inpainting that intensify with increasing resolution and texture complexity. Patch-Adapter leverages a two-stage adapter architecture to scale the diffusion model's resolution from 1K to 4K+ without requiring structural overhauls: (1) Dual Context Adapter learns coherence between masked and unmasked regions at reduced resolutions to establish global structural consistency; and (2) Reference Patch Adapter implements a patch-level attention mechanism for full-resolution inpainting, preserving local detail fidelity through adaptive feature fusion. This dual-stage architecture uniquely addresses the scalability gap in high-resolution inpainting by decoupling global semantics from localized refinement. Experiments demonstrate that Patch-Adapter not only resolves artifacts common in large-scale inpainting but also achieves state-of-the-art performance on the OpenImages and Photo-Concept-Bucket datasets, outperforming existing methods in both perceptual quality and text-prompt adherence.

InkBrush is a new sketch-based 3D drawing tool for creating 3D ink paintings using free-form 3D ink strokes. It offers a digital calligraphy brush and various editing tools to generate realistic ink-like brush strokes with attributes like hairy edges, ink drips, and scattered dots. Users can adjust parameters such as moisture, color, darkness, dryness, and stroke style to customize the appearance of the brush strokes. The development of InkBrush was guided by a design study involving artists and designers. It was developed as a plugin for Blender, a popular 3D modeling tool, and its effectiveness and usability were evaluated through a user study involving 75 participants. Preliminary feedback from the participants was overwhelmingly positive, indicating that InkBrush was intuitive and easy to use. Following this, we also sought in-depth assessments from experts in ink painting and 3D design. Their evaluations further demonstrated the effectiveness of InkBrush.

To improve meta-generalization, i.e., accommodating out-of-domain meta-testing tasks beyond meta-training ones, is of significance to extending the success of meta-learning beyond standard benchmarks. Previous heterogeneous meta-learning algorithms have shown that tailoring the global meta-knowledge by the learned clusters during meta-training promotes better meta-generalization to novel meta-testing tasks. Inspired by this, we propose a novel multi-objective perspective to sharpen the compositionality of the meta-trained clusters, through which we have empirically validated that the meta-generalization further improves. Grounded on the hierarchically structured meta-learning framework, we formulate a hypervolume loss to evaluate the degree of conflict between multiple cluster-conditioned parameters in the two-dimensional loss space over two randomly chosen tasks belonging to two clusters and two mixed tasks imitating out-of-domain tasks. Experimental results on more than 16 few-shot image classification datasets show not only improved performance on out-of-domain meta-testing datasets but also better clusters in visualization.

Shadow puppetry, a culturally rich storytelling art, faces challenges transitioning to the digital realm. Creators in the early design phase struggle with crafting intricate patterns, textures, and basic animations while adhering to stylistic conventions - hindering creativity, especially for novices. This paper presents Lumina, a tool to facilitate the early Chinese shadow puppet design stage. Lumina provides contour templates, animations, scene editing tools, and machine-generated traditional puppet patterns. These features liberate creators from tedious tasks, allowing focus on the creative process. Developed based on a formative study with puppet creators, the web-based Lumina enables wide dissemination. An evaluation with 18 participants demonstrated Lumina’s effectiveness and ease of use, with participants successfully creating designs spanning traditional themes to contemporary and science-fiction concepts.

With the advancement of interactive technology in the information age, the problem of “visual blindness” in the field of display design has become increasingly prevalent in public scene interaction design. Therefore, it has become crucial to address how new forms of interaction and interaction scenes can be adopted to attract the public. In the context of robotics’ continuous development, robots are playing an increasingly prominent role as interactive subjects in public scene interaction experiences. In new fields such as digital entertainment, spatial experience, and new media art, various typical scenes of robot interaction have emerged. In this study, a window robot named “Yousu” was developed based on an ancient Chinese mythological character and deployed in the window of a bookstore in Beijing. A user experiment was conducted to investigate how to design a reasonable and effective character robot interaction in public scenes to enhance the interaction scenes’ attractiveness.

The rapid pace of malware development and the widespread use of code obfuscation, polymorphism, and morphing techniques pose a considerable challenge to detecting and analyzing malware. Today, it is difficult for antivirus applications to use traditional signature-based detection methods to detect morphing malware. Thus, the emergence of structure graph-based detection methods has become a hope to solve this challenge. In this work, we propose a method for detecting malware using graphs' spectral heat and wave signatures, which are efficient and size- and permutation-invariant. We extracted 250 and 1,000 heat and wave representations, and we trained and tested heat and wave representations on eight machine learning classifiers. We used a dataset of 37,537 unpacked Windows malware executables and extracted the control flow graph (CFG) of each windows malware to obtain the spectral representations. Our experimental results showed that by using heat and wave spectral graph theory, the best malware analysis accuracy reached 95.9%.

Large DNNs with mixed-precision quantization can achieve ultra-high compression while retaining high classification performance. However, because of the challenges in finding an accurate metric that can guide the optimization process, these methods either sacrifice significant performance compared to the 32-bit floating-point (FP-32) baseline or rely on a compute-expensive, iterative training policy that requires the availability of a pre-trained baseline. To address this issue, this paper presents BMPQ, a training method that uses bit gradients to analyze layer sensitivities and yield mixed-precision quantized models. BMPQ requires a single training iteration but does not need a pre-trained baseline. It uses an integer linear program (ILP) to dynamically adjust the precision of layers during training, subject to a fixed hardware budget. To evaluate the efficacy of BMPQ, we conduct extensive experiments with VGG16 and ResNet18 on CIFAR-10, CIFAR-100, and Tiny-ImageNet datasets. Compared to the baseline FP-32 models, BMPQ can yield models that have 15.4x fewer parameter bits with negligible drop in accuracy. Compared to the SOTA “during training”, mixed-precision training scheme, our models are 2.1 x, 2.2x, and 2.9x smaller, on CIFAR-10, CIFAR-100, and Tiny-ImageNet, respectively, with an improved accuracy of up to 14.54%.

Knowledge distillation (KD) has recently been identified as a method that can unintentionally leak private information regarding the details of a teacher model to an unauthorized student. Recent research in developing undistillable nasty teachers that can protect model confidentiality has gained significant attention. However, the level of protection these nasty models offer has been largely untested. In this paper, we show that transferring knowledge to a shallow sub-section of a student can largely reduce a teacher’s influence. By exploring the depth of the shallow subsection, we then present a distillation technique that enables a skeptical student model to learn even from a nasty teacher. To evaluate the efficacy of our skeptical students, we conducted experiments with several models with KD on both training data-available and data-free scenarios for various datasets. While distilling from nasty teachers, compared to the normal student models, skeptical students consistently provide superior classification performance of up to ∼59.5%. Moreover, similar to normal students, skeptical students maintain high classification accuracy when distilled from a normal teacher, showing their efficacy irrespective of the teacher being nasty or not. We believe the ability of skeptical students to largely diminish the KD-immunity of potentially nasty teachers will motivate the research community to create more robust mechanisms for model confidentiality. We have open-sourced the code at https://github.com/ksouvik52/Skeptical2021

The multiple traveling salesman problem (MTSP) is an important combinatorial optimization problem. It has been widely and successfully applied to the practical cases in which multiple traveling individuals (salesmen) share the common workspace (city set). However, it cannot represent some application problems where multiple traveling individuals not only have their own exclusive tasks but also share a group of tasks with each other. This work proposes a new MTSP called colored traveling salesman problem (CTSP) for handling such cases. Two types of city groups are defined, i.e., each group of exclusive cities of a single color for a salesman to visit and a group of shared cities of multiple colors allowing all salesmen to visit. Evidences show that CTSP is NP-hard and a multidepot MTSP and multiple single traveling salesman problems are its special cases. We present a genetic algorithm (GA) with dual-chromosome coding for CTSP and analyze the corresponding solution space. Then, GA is improved by incorporating greedy, hill-climbing (HC), and simulated annealing (SA) operations to achieve better performance. By experiments, the limitation of the exact solution method is revealed and the performance of the presented GAs is compared. The results suggest that SAGA can achieve the best quality of solutions and HCGA should be the choice making good tradeoff between the solution quality and computing time.

This work formulates for the first time a multiple traveling salesman problem (MTSP) with ordinary and exclusive cities, denoted by MTSP for short. In the original MTSP, a city can be visited by any traveling salesman and is thus renamed as an ordinary one in MTSP. A new class of cities is introduced in MTSP, called exclusive ones. They are divided into groups, each of which can be exclusively visited by a specified or predetermined salesman. To solve MTSP, a genetic algorithm is presented. It encodes cities and salesman into two single chromosomes. Accordingly, three modes of crossover and mutation operators are designed, i.e., simple city crossover and mutation (CCM), simple salesman crossover and mutation, and mixed city-salesman crossover and mutation. All the operations of crossover and mutation follow the proper relationship between cities and salesman. With the help of an MTSP example, the performance of the proposed algorithm with three modes of crossover and mutation operators is compared and analyzed. The simulation results show that the algorithm can solve MTSP with rapid convergence with CCM being the best mode of the operators.