Antimicrobial peptides (AMPs) are promising alternatives to conventional antibiotics, yet early translation is often hindered by the perceived coupling between antibacterial potency and mammalian toxicity. This assumption complicates prioritization: highly active candidates are frequently suspected to be hemolytic, while existing multi-task predictors rarely reveal where selectivity resides in sequence space. Here, we present Multifunctional AMP Learning Engine (MAPLE), an interpretable dual-stream framework for AMP identification and systematic category-specific functional profiling across 14 activity categories directly from peptide sequences. MAPLE combines protein language model embeddings with explicit physicochemical descriptors, enabling robust task-specific prediction under severe label imbalance. Across the benchmark dataset and a sequence-non-overlapping independent validation set, MAPLE achieves consistently well-balanced performance, including on low-prevalence but clinically relevant endpoints. Building on this predictive basis, we conduct systematic k-mer enrichment to map motif-level selectivity and show that potency-hemolysis coupling is motif-regime-dependent rather than universal. Motifs most strongly enriched for antibacterial activity exhibit reduced hemolytic overlap and occupy a physicochemical regime characterized by moderate cationicity, lower hydrophobicity, and higher amphipathicity. We further provide a proof-of-concept prioritization workflow leveraging antibacterial-selective motifs, with structural modeling yielding conformations consistent with amphipathic α-helices. Despite limitations of predominantly binary annotations and incomplete structural integration, MAPLE offers reproducible sequence-level hypotheses and prioritization principles to support the engineering of potent and safer AMPs.

Abstract Existing image restoration methods primarily rely on the posterior distribution of natural images but are often limited by their dependence on known degradations and supervised training. To this end, we propose Generative Diffusion Prior (GDP), an unsupervised sampling-based framework that effectively models posterior distributions for image and video restoration. GDP utilizes a single pre-trained denoising diffusion probabilistic model (DDPM) to solve a wide range of linear, non-linear, and blind inverse problems without explicit degradation assumptions. Specifically, GDP systematically explores a conditional guidance protocol, which proves more practical and effective than conventional methods of adding guidance. Furthermore, GDP incorporates a degradation model optimization mechanism during the denoising process, enabling blind image restoration. Besides, we introduce a patch-based strategy, allowing GDP to handle images of arbitrary resolution. We extensively evaluate GDP on multiple image and video restoration tasks, including super-resolution, deblurring, inpainting, and colorization, as well as more challenging applications such as low-light enhancement, HDR recovery, and LDR video enhancement. Experimental results demonstrate that GDP outperforms leading unsupervised methods across diverse benchmarks in both reconstruction accuracy and perceptual quality, while demonstrating robust generalization to images and videos of any size. Our project page at https://generativediffusionprior.github.io/.

Electric power is the foundation of modern society, yet Europe is currently facing an energy crisis, increasing interest in power generation, energy infrastructure, and grid resilience. However, power plant data are complex and multidimensional, making it difficult to gain an overview or understanding. Visualization methods can help to reduce cognitive load and facilitate exploration of such data. In this paper, we propose EuroEnergyVis, a web-based visualization approach designed for the interactive exploration of power plant data across European countries. The design requirements were motivated by gaps identified in prior work. We conducted interviews with six domain experts in power systems and energy, which indicate that our tool enhances the user experience when exploring European power plants. Their reflections also suggest directions for future work.

We introduce MVTexGen, a novel method for generating textures on 3D geometries using a 2D text-to-image diffusion model. Traditional project-and-inpaint techniques, often result in texture inconsistencies due to uneven diffusion across views. We address this issue by integration of a multi-view prior into the generation process, ensuring synchronous view generation and uniformity in overlapping areas. It combines Multi-View Diffusion models with depth-conditioned diffusion models to create consistent, depth-aware texture maps. Addressing latent space gaps, MVTexGen refines the texture map by increasing view count and fusing denoised views for uniformity. Our extensive experiments on benchmark datasets show MVTexGen’s superiority in generating high-quality, detailed textures, outperforming current state-of-the-art methods.

The emergence of more and more mobile applications in recent years has driven the development of application recommendation algorithms. However, many algorithms are limited to using users’ ratings for apps while considering other relevant information as marginal. In this paper, we consider the user’s app usage time and the user’s forum level in addition to the rating, explore the impact of different attributes and multidimensional attributes on the recommendation algorithms, and propose a model MIAE (Multi-info. Autoencoders for Recommendation) based on a Neural Tangent Kernel(NTK) that can accommodate multiple information dimensions. Specifically, our approach involves utilizing a fully-connected neural network with NTK parameterization. We introduce ridge regression as a regularization technique to convert the recommendation problem into a probabilistic framework. To obtain results, we employ a gradient descent method based on the NTK, enabling efficient optimization. We tested our approach using both comparison experiment and ablation experiment. In the comparison experiment, we used two representative algorithms, GLocal-K [1] and CosRec [2], as baselines. Quantitative results on our crawled taptap dataset show that our proposed NTK-based approach, MIAE, with one-dimensional information input outperforms representative algorithms, achieving high performance on various evaluation metrics. In our ablation experiment, we examined several variables, including the user’s rating of the app, app usage duration, and forum level. The results demonstrated that utilizing the individual variable ‘Level’ led to the highest recall and F1 scores. On the other hand, employing a combination of variables such as ‘Level + Duration’ resulted in the best precision and NDCG scores. This finding confirms the beneficial impact of both the user’s app usage duration and level on app recommendations.

Faced with so many mobile applications in the app store, users have difficulties finding their preferred mobile applications. Existing studies do not comprehensively consider implicit feedback in mobile applications and thus do not combine behavioral information and published information together to make recommendations. This paper proposes a novel method to recommend mobile applications based on metagraph embedding using the combination of behavioral information and published information. Specifically, this paper constructed a conceptual model using the combinations of behavioral information and published information that could well portray users and mobile applications. Based on this conceptual model, six metagraphs are designed to interpret the multidimensional relationships between users and mobile applications in the model. By random walking guided by each metagraph, a series of node sequences that could express node neighborhood are obtained. Finally, the similarity between users and apps is calculated using the embedded vector of each node, and the recommendations are given to the user. Based on a real-world dataset, we evaluate the performance of our method. The experimental result shows that our method outperforms existing models and methods in all metrics, in which the average F1-measure increases by 19.21%, and the average NDCG increases by 4.99%.

Mesh generation is of great value in various applications involving computer graphics and virtual content, yet designing generative models for meshes is challenging due to their irregular data structure and inconsistent topology of meshes in the same category. In this work, we design a novel sparse latent point diffusion model for mesh generation. Our key insight is to regard point clouds as an intermediate representation of meshes, and model the distribution of point clouds instead. While meshes can be generated from point clouds via techniques like Shape as Points (SAP), the challenges of directly generating meshes can be effectively avoided. To boost the efficiency and controllability of our mesh generation method, we propose to further encode point clouds to a set of sparse latent points with pointwise semantic meaningful features, where two DDPMs are trained in the space of sparse latent points to respectively model the distribution of the latent point positions and features at these latent points. We find that sampling in this latent space is faster than directly sampling dense point clouds. Moreover, the sparse latent points also enable us to explicitly control both the overall structures and local details of the generated meshes. Extensive experiments are conducted on the ShapeNet dataset, where our proposed sparse latent point diffusion model achieves superior performance in terms of generation quality and controllability when compared to existing methods. Project page, code and appendix: https://slide-3d.github.io.

We consider the fundamental problem of deriving quantitative bounds on the probability that a given assertion is violated in a probabilistic program. We provide automated algorithms that obtain both lower and upper bounds on the assertion violation probability. The main novelty of our approach is that we prove new and dedicated fixed-point theorems which serve as the theoretical basis of our algorithms and enable us to reason about assertion violation bounds in terms of pre and post fixed-point functions. To synthesize such fixed-points, we devise algorithms that utilize a wide range of mathematical tools, including repulsing ranking supermartingales, Hoeffding's lemma, Minkowski decompositions, Jensen's inequality, and convex optimization. On the theoretical side, we provide (i) the first automated algorithm for lower-bounds on assertion violation probabilities, (ii) the first complete algorithm for upper-bounds of exponential form in affine programs, and (iii) provably and significantly tighter upper-bounds than the previous approaches. On the practical side, we show our algorithms can handle a wide variety of programs from the literature and synthesize bounds that are remarkably tighter than previous results, in some cases by thousands of orders of magnitude.

This paper is an introduction to the "Future IKEA Catalogue", enclosed here as an example of a design fiction produced from a long standing industrial-academic collaboration. We introduce the catalogue here by discussing some of our experiences using design fiction` with companies and public sector bodies, giving some background to the catalogue and the collaboration which produced it. We have found design fiction to be a useful tool to support collaboration with industrial partners in research projects - it provides a way of thinking and talking about present day concepts, and present day constraints, without being overly concerned with contemporary challenges, or the requirements of academic validation. In particular, there are two main aspects of this we will discuss here, aspects that are visible in the enclosed catalogue itself. The first is the potential of design fiction as a sort of 'boundary object' in industry and academic collaboration, and second the role of critique. After this introduction to the paper we enclose the output of our collaboration in the form of the catalogue itself.

In order to invent and investigate new approaches for the use of enjoying live video, we suggest a combination of emerging mobile webcasting with artistic ambient video, which would enable a form of user generated broadcasts from individually selected cherished places for home decoration. Drawing on the approach of Research through Design we present a study of people who have occasional access to highly appreciated geographical locations, a design instantiation and prototype called LiveNature, as well as a system implementation. We present the result of a technical evaluation, which was conducted during two weeks of deployment. It shows that mobile webcasting provide continuous and stable streams of such a quality that it can be presented for home decoration, and that the video can be combined with real time sensor data to generate aesthetically interesting hybrid media. We also learned that the use of mobile webcasting for home decoration raises new challenges in order to provide unobtrusive and glance based interaction.