Search

The fields of 3D reconstruction and text-based 3D editing have advanced significantly with the evolution of text-based diffusion models. While existing 3D editing methods excel at modifying color, texture, and style, they struggle with extensive geometric or appearance changes, thus limiting their applications. We propose Perturb-and-Revise, which makes possible a variety of NeRF editing. First, we perturb the NeRF parameters with random initializations to create a versatile initialization. We automatically determine the perturbation magnitude through analysis of the local loss landscape. Then, we revise the edited NeRF via generative trajectories. Combined with the generative process, we impose identity-preserving gradients to refine the edited NeRF. Extensive experiments demonstrate that Perturb-and-Revise facilitates flexible, effective, and consistent editing of color, appearance, and geometry in 3D. For 360{\deg} results, please visit our project page: https://susunghong.github.io/Perturb-and-Revise., Comment: Project page: https://susunghong.github.io/Perturb-and-Revise

We introduce Edify Image, a family of diffusion models capable of generating photorealistic image content with pixel-perfect accuracy. Edify Image utilizes cascaded pixel-space diffusion models trained using a novel Laplacian diffusion process, in which image signals at different frequency bands are attenuated at varying rates. Edify Image supports a wide range of applications, including text-to-image synthesis, 4K upsampling, ControlNets, 360 HDR panorama generation, and finetuning for image customization.

We present Fashion-VDM, a video diffusion model (VDM) for generating virtual try-on videos. Given an input garment image and person video, our method aims to generate a high-quality try-on video of the person wearing the given garment, while preserving the person's identity and motion. Image-based virtual try-on has shown impressive results; however, existing video virtual try-on (VVT) methods are still lacking garment details and temporal consistency. To address these issues, we propose a diffusion-based architecture for video virtual try-on, split classifier-free guidance for increased control over the conditioning inputs, and a progressive temporal training strategy for single-pass 64-frame, 512px video generation. We also demonstrate the effectiveness of joint image-video training for video try-on, especially when video data is limited. Our qualitative and quantitative experiments show that our approach sets the new state-of-the-art for video virtual try-on. For additional results, visit our project page: https://johannakarras.github.io/Fashion-VDM., Comment: Accepted to SIGGRAPH Asia 2024

The distributed data infrastructure in Internet of Things (IoT) ecosystems requires efficient data-series compression methods, along with the ability to feed different accuracy demands. However, the compression performance of existing compression methods degrades sharply when calling for ultra-accurate data recovery. In this paper, we introduce SHRINK, a novel highly accurate data compression method that offers a higher compression ratio and also lower runtime than prior compressors. SHRINK extracts data semantics in the form of linear segments to construct a compact knowledge base, using a dynamic error threshold that it adapts to data characteristics. Then, it captures the remaining data details as residuals to support lossy compression at diverse resolutions as well as lossless compression. As SHRINK identifies repeated semantics, its compression ratio increases with data size. Our experimental evaluation demonstrates that SHRINK outperforms state-of-art methods with an up to threefold improvement in compression ratio., Comment: 11 pages

This article presents a Visual Servoing Nonlinear Model Predictive Control (NMPC) scheme for autonomously tracking a moving target using multirotor Unmanned Aerial Vehicles (UAVs). The scheme is developed for surveillance and tracking of contour-based areas with evolving features. NMPC is used to manage input and state constraints, while additional barrier functions are incorporated in order to ensure system safety and optimal performance. The proposed control scheme is designed based on the extraction and implementation of the full dynamic model of the features describing the target and the state variables. Real-time simulations and experiments using a quadrotor UAV equipped with a camera demonstrate the effectiveness of the proposed strategy.

How can we mine frequent path regularities from a graph with edge labels and vertex attributes? The task of association rule mining successfully discovers regular patterns in item sets and substructures. Still, to our best knowledge, this concept has not yet been extended to path patterns in large property graphs. In this paper, we introduce the problem of path association rule mining (PARM). Applied to any \emph{reachability path} between two vertices within a large graph, PARM discovers regular ways in which path patterns, identified by vertex attributes and edge labels, co-occur with each other. We develop an efficient and scalable algorithm PIONEER that exploits an anti-monotonicity property to effectively prune the search space. Further, we devise approximation techniques and employ parallelization to achieve scalable path association rule mining. Our experimental study using real-world graph data verifies the significance of path association rules and the efficiency of our solutions.

Authoring survey or review articles still requires significant tedious manual effort, despite many advancements in research knowledge management having the potential to improve efficiency, reproducibility, and reuse. However, these advancements bring forth an increasing number of approaches, tools, and systems, which often cover only specific stages and lack a comprehensive workflow utilizing their task-specific strengths. We propose the Streamlined Workflow Automation for Machine-actionable Systematic Literature Reviews (SWARM-SLR) to crowdsource the improvement of SLR efficiency while maintaining scientific integrity in a state-of-the-art knowledge discovery and distribution process. The workflow aims to domain-independently support researchers in collaboratively and sustainably managing the rising scholarly knowledge corpus. By synthesizing guidelines from the literature, we have composed a set of 65 requirements, spanning from planning to reporting a review. Existing tools were assessed against these requirements and synthesized into the SWARM-SLR workflow prototype, a ready-for-operation software support tool. The SWARM-SLR was evaluated via two online surveys, which largely confirmed the validity of the 65 requirements and situated 11 tools to the different life-cycle stages. The SWARM-SLR workflow was similarly evaluated and found to be supporting almost the entire span of an SLR, excelling specifically in search and retrieval, information extraction, knowledge synthesis, and distribution. Our SWARM-SLR requirements and workflow support tool streamlines the SLR support for researchers, allowing sustainable collaboration by linking individual efficiency improvements to crowdsourced knowledge management. If these efforts are continued, we expect the increasing number of tools to be manageable and usable inside fully structured, (semi-)automated literature review workflows., Comment: This preprint has not undergone peer review (when applicable) or any post-submission improvements or corrections

Angle-resolved photoemission spectroscopy (ARPES) is a powerful experimental technique to determine the electronic structure of solids. Advances in light sources for ARPES experiments are currently leading to a vast increase of data acquisition rates and data quantity. On the other hand, access time to the most advanced ARPES instruments remains strictly limited, calling for fast, effective, and on-the-fly data analysis tools to exploit this time. In response to this need, we introduce ARPESNet, a versatile autoencoder network that efficiently summmarises and compresses ARPES datasets. We train ARPESNet on a large and varied dataset of 2-dimensional ARPES data extracted by cutting standard 3-dimensional ARPES datasets along random directions in $\mathbf{k}$. To test the data representation capacity of ARPESNet, we compare $k$-means clustering quality between data compressed by ARPESNet, data compressed by discrete cosine transform, and raw data, at different noise levels. ARPESNet data excels in clustering quality despite its high compression ratio.

The primary axes of interest in image-generating diffusion models are image quality, the amount of variation in the results, and how well the results align with a given condition, e.g., a class label or a text prompt. The popular classifier-free guidance approach uses an unconditional model to guide a conditional model, leading to simultaneously better prompt alignment and higher-quality images at the cost of reduced variation. These effects seem inherently entangled, and thus hard to control. We make the surprising observation that it is possible to obtain disentangled control over image quality without compromising the amount of variation by guiding generation using a smaller, less-trained version of the model itself rather than an unconditional model. This leads to significant improvements in ImageNet generation, setting record FIDs of 1.01 for 64x64 and 1.25 for 512x512, using publicly available networks. Furthermore, the method is also applicable to unconditional diffusion models, drastically improving their quality., Comment: NeurIPS 2024

In the last two decades, several researchers provided snapshots of the "current" state and evolution of empirical research in requirements engineering (RE) through literature reviews. However, these literature reviews were not sustainable, as none built on or updated previous works due to the unavailability of the extracted and analyzed data. KG-EmpiRE is a Knowledge Graph (KG) of empirical research in RE based on scientific data extracted from currently 680 papers published in the IEEE International Requirements Engineering Conference (1994-2022). KG-EmpiRE is maintained in the Open Research Knowledge Graph (ORKG), making all data openly and long-term available according to the FAIR data principles. Our long-term goal is to constantly maintain KG-EmpiRE with the research community to synthesize a comprehensive, up-to-date, and long-term available overview of the state and evolution of empirical research in RE. Besides KG-EmpiRE, we provide its analysis with all supplementary materials in a repository. This repository contains all files with instructions for replicating and (re-)using the analysis locally or via executable environments and for repeating the research approach. Since its first release based on 199 papers (2014-2022), KG-EmpiRE and its analysis have been updated twice, currently covering over 650 papers. KG-EmpiRE and its analysis demonstrate how innovative infrastructures, such as the ORKG, can be leveraged to make data from literature reviews FAIR, openly available, and maintainable for the research community in the long term. In this way, we can enable replicable, (re-)usable, and thus sustainable literature reviews to ensure the quality, reliability, and timeliness of their research results., Comment: Accepted for publication at the 32nd IEEE International Requirements Engineering conference (RE) 2024

We consider a problem of placing generators of rewards to be collected by randomly moving agents in a network. In many settings, the precise mobility pattern may be one of several possible, based on parameters outside our control, such as weather conditions. The placement should be robust to this uncertainty, to gain a competent total reward across possible networks. To study such scenarios, we introduce the Robust Reward Placement problem (RRP). Agents move randomly by a Markovian Mobility Model with a predetermined set of locations whose connectivity is chosen adversarially from a known set $\Pi$ of candidates. We aim to select a set of reward states within a budget that maximizes the minimum ratio, among all candidates in $\Pi$, of the collected total reward over the optimal collectable reward under the same candidate. We prove that RRP is NP-hard and inapproximable, and develop $\Psi$-Saturate, a pseudo-polynomial time algorithm that achieves an $\epsilon$-additive approximation by exceeding the budget constraint by a factor that scales as $O(\ln |\Pi|/\epsilon)$. In addition, we present several heuristics, most prominently one inspired by a dynamic programming algorithm for the max-min 0-1 KNAPSACK problem. We corroborate our theoretical analysis with an experimental evaluation on synthetic and real data., Comment: Accepted for publication in IJCAI 2024

Can we assess a priori how well a knowledge graph embedding will perform on a specific downstream task and in a specific part of the knowledge graph? Knowledge graph embeddings (KGEs) represent entities (e.g., "da Vinci," "Mona Lisa") and relationships (e.g., "painted") of a knowledge graph (KG) as vectors. KGEs are generated by optimizing an embedding score, which assesses whether a triple (e.g., "da Vinci," "painted," "Mona Lisa") exists in the graph. KGEs have been proven effective in a variety of web-related downstream tasks, including, for instance, predicting relationships among entities. However, the problem of anticipating the performance of a given KGE in a certain downstream task and locally to a specific individual triple, has not been tackled so far. In this paper, we fill this gap with ReliK, a Reliability measure for KGEs. ReliK relies solely on KGE embedding scores, is task- and KGE-agnostic, and requires no further KGE training. As such, it is particularly appealing for semantic web applications which call for testing multiple KGE methods on various parts of the KG and on each individual downstream task. Through extensive experiments, we attest that ReliK correlates well with both common downstream tasks, such as tail or relation prediction and triple classification, as well as advanced downstream tasks, such as rule mining and question answering, while preserving locality.

The Moran process is a classic stochastic process that models the rise and takeover of novel traits in network-structured populations. In biological terms, a set of mutants, each with fitness $m\in(0,\infty)$ invade a population of residents with fitness $1$. Each agent reproduces at a rate proportional to its fitness and each offspring replaces a random network neighbor. The process ends when the mutants either fixate (take over the whole population) or go extinct. The fixation probability measures the success of the invasion. To account for environmental heterogeneity, we study a generalization of the Standard process, called the Heterogeneous Moran process. Here, the fitness of each agent is determined both by its type (resident/mutant) and the node it occupies. We study the natural optimization problem of seed selection: given a budget $k$, which $k$ agents should initiate the mutant invasion to maximize the fixation probability? We show that the problem is strongly inapproximable: it is $\mathbf{NP}$-hard to distinguish between maximum fixation probability 0 and 1. We then focus on mutant-biased networks, where each node exhibits at least as large mutant fitness as resident fitness. We show that the problem remains $\mathbf{NP}$-hard, but the fixation probability becomes submodular, and thus the optimization problem admits a greedy $(1-1/e)$-approximation. An experimental evaluation of the greedy algorithm along with various heuristics on real-world data sets corroborates our results., Comment: Accepted for publication in IJCAI 2024

Guidance is a crucial technique for extracting the best performance out of image-generating diffusion models. Traditionally, a constant guidance weight has been applied throughout the sampling chain of an image. We show that guidance is clearly harmful toward the beginning of the chain (high noise levels), largely unnecessary toward the end (low noise levels), and only beneficial in the middle. We thus restrict it to a specific range of noise levels, improving both the inference speed and result quality. This limited guidance interval improves the record FID in ImageNet-512 significantly, from 1.81 to 1.40. We show that it is quantitatively and qualitatively beneficial across different sampler parameters, network architectures, and datasets, including the large-scale setting of Stable Diffusion XL. We thus suggest exposing the guidance interval as a hyperparameter in all diffusion models that use guidance., Comment: NeurIPS 2024

Angle-resolved photoemission spectroscopy (ARPES) is a technique used to map the occupied electronic structure of solids. Recent progress in X-ray focusing optics has led to the development of ARPES into a microscopic tool, permitting the electronic structure to be spatially mapped across the surface of a sample. This comes at the expense of a time-consuming scanning process to cover not only a three-dimensional energy-momentum ($E, k_z, k_y$) space but also the two-dimensional surface area. Here, we implement a protocol to autonomously search both $\mathbf{k}$- and real space in order to find positions of particular interest, either because of their high photoemission intensity or because of sharp spectral features. The search is based on the use of Gaussian process regression and can easily be expanded to include additional parameters or optimisation criteria. This autonomous experimental control is implemented on the SGM4 micro-focus beamline of the synchrotron radiation source ASTRID2.

The need to identify graphs having small structural distance from a query arises in biology, chemistry, recommender systems, and social network analysis. Among several methods to measure inter graph distance, Graph Edit Distance (GED) is preferred for its comprehensibility, yet hindered by the NP-hardness of its computation. State-of-the-art GED approximations predominantly employ neural methods, which, however, (i) lack an explanatory edit path corresponding to the approximated GED; (ii) require the NP-hard generation of ground-truth GEDs for training; and (iii) necessitate separate training on each dataset. In this paper, we propose an efficient algebraic unsuper vised method, EUGENE, that approximates GED and yields edit paths corresponding to the approx imated cost, while eliminating the need for ground truth generation and data-specific training. Extensive experimental evaluation demonstrates that the aforementioned benefits of EUGENE do not come at the cost of efficacy. Specifically, EUGENE consistently ranks among the most accurate methods across all of the benchmark datasets and outperforms majority of the neural approaches.

Engineering sciences, such as energy system research, play an important role in developing solutions to technical, environmental, economic, and social challenges of our modern society. In this context, the transformation of energy systems into climate-neutral systems is one of the key strategies for mitigating climate change. For the transformation of energy systems, engineers model, simulate and analyze scenarios and transformation pathways to initiate debates about possible transformation strategies. For these debates and research in general, all steps of the research process must be traceable to guarantee the trustworthiness of published results, avoid redundancies, and ensure their social acceptance. However, the analysis of energy systems is an interdisciplinary field as the investigations of large, complex energy systems often require the use of different software applications and large amounts of heterogeneous data. Engineers must therefore communicate, understand, and (re)use heterogeneous scientific knowledge and data. Although the importance of FAIR scientific knowledge and data in the engineering sciences and energy system research is increasing, little research has been conducted on this topic. When it comes to publishing scientific knowledge and data from publications, software, and datasets (such as models, scenarios, and simulations) openly available and transparent, energy system research lags behind other research domains. According to Schmitt et al. and Nie{\ss}e et al., engineers need technical support in the form of infrastructures, services, and terminologies to improve communication, understanding, and (re)use of scientific knowledge and data., Comment: 1. NFDI4Energy Conference

Diffusion models currently dominate the field of data-driven image synthesis with their unparalleled scaling to large datasets. In this paper, we identify and rectify several causes for uneven and ineffective training in the popular ADM diffusion model architecture, without altering its high-level structure. Observing uncontrolled magnitude changes and imbalances in both the network activations and weights over the course of training, we redesign the network layers to preserve activation, weight, and update magnitudes on expectation. We find that systematic application of this philosophy eliminates the observed drifts and imbalances, resulting in considerably better networks at equal computational complexity. Our modifications improve the previous record FID of 2.41 in ImageNet-512 synthesis to 1.81, achieved using fast deterministic sampling. As an independent contribution, we present a method for setting the exponential moving average (EMA) parameters post-hoc, i.e., after completing the training run. This allows precise tuning of EMA length without the cost of performing several training runs, and reveals its surprising interactions with network architecture, training time, and guidance.

[Background.] Empirical research in requirements engineering (RE) is a constantly evolving topic, with a growing number of publications. Several papers address this topic using literature reviews to provide a snapshot of its "current" state and evolution. However, these papers have never built on or updated earlier ones, resulting in overlap and redundancy. The underlying problem is the unavailability of data from earlier works. Researchers need technical infrastructures to conduct sustainable literature reviews. [Aims.] We examine the use of the Open Research Knowledge Graph (ORKG) as such an infrastructure to build and publish an initial Knowledge Graph of Empirical research in RE (KG-EmpiRE) whose data is openly available. Our long-term goal is to continuously maintain KG-EmpiRE with the research community to synthesize a comprehensive, up-to-date, and long-term available overview of the state and evolution of empirical research in RE. [Method.] We conduct a literature review using the ORKG to build and publish KG-EmpiRE which we evaluate against competency questions derived from a published vision of empirical research in software (requirements) engineering for 2020 - 2025. [Results.] From 570 papers of the IEEE International Requirements Engineering Conference (2000 - 2022), we extract and analyze data on the reported empirical research and answer 16 out of 77 competency questions. These answers show a positive development towards the vision, but also the need for future improvements. [Conclusions.] The ORKG is a ready-to-use and advanced infrastructure to organize data from literature reviews as knowledge graphs. The resulting knowledge graphs make the data openly available and maintainable by research communities, enabling sustainable literature reviews., Comment: Accepted for publication at the 17th ACM/IEEE International Symposium on Empirical Software Engineering and Measurement (ESEM 2023)

The reuse of research software is central to research efficiency and academic exchange. The application of software enables researchers with varied backgrounds to reproduce, validate, and expand upon study findings. Furthermore, the analysis of open source code aids in the comprehension, comparison, and integration of approaches. Often, however, no further use occurs because relevant software cannot be found or is incompatible with existing research processes. This results in repetitive software development, which impedes the advancement of individual researchers and entire research communities. In this article, the DataDesc ecosystem is presented, an approach to describing data models of software interfaces with detailed and machine-actionable metadata. In addition to a specialized metadata schema, an exchange format and support tools for easy collection and the automated publishing of software documentation are introduced. This approach practically increases the FAIRness, i.e., findability, accessibility, interoperability, and so the reusability of research software, as well as effectively promotes its impact on research.

We present DreamPose, a diffusion-based method for generating animated fashion videos from still images. Given an image and a sequence of human body poses, our method synthesizes a video containing both human and fabric motion. To achieve this, we transform a pretrained text-to-image model (Stable Diffusion) into a pose-and-image guided video synthesis model, using a novel fine-tuning strategy, a set of architectural changes to support the added conditioning signals, and techniques to encourage temporal consistency. We fine-tune on a collection of fashion videos from the UBC Fashion dataset. We evaluate our method on a variety of clothing styles and poses, and demonstrate that our method produces state-of-the-art results on fashion video animation.Video results are available on our project page., Comment: Project page: https://grail.cs.washington.edu/projects/dreampose/

Scientific knowledge graphs have been proposed as a solution to structure the content of research publications in a machine-actionable way and enable more efficient, computer-assisted workflows for many research activities. Crowd-sourcing approaches are used frequently to build and maintain such scientific knowledge graphs. To contribute to scientific knowledge graphs, researchers need simple and easy-to-use solutions to generate new knowledge graph elements and establish the practice of semantic representations in scientific communication. In this paper, we present a workflow for authors of scientific documents to specify their contributions with a LaTeX package, called SciKGTeX, and upload them to a scientific knowledge graph. The SciKGTeX package allows authors of scientific publications to mark the main contributions of their work directly in LaTeX source files. The package embeds marked contributions as metadata into the generated PDF document, from where they can be extracted automatically and imported into a scientific knowledge graph, such as the ORKG. This workflow is simpler and faster than current approaches, which make use of external web interfaces for data entry. Our user evaluation shows that SciKGTeX is easy to use, with a score of 79 out of 100 on the System Usability Scale, as participants of the study needed only 7 minutes on average to annotate the main contributions on a sample abstract of a published paper. Further testing shows that the embedded contributions can be successfully uploaded to ORKG within ten seconds. SciKGTeX simplifies the process of manual semantic annotation of research contributions in scientific articles. Our workflow demonstrates how a scientific knowledge graph can automatically ingest research contributions from document metadata., Comment: Accepted for publication at the ACM/IEEE Joint Conference on Digital Libraries 2023 (JCDL2023)

We present a diffusion-based model for 3D-aware generative novel view synthesis from as few as a single input image. Our model samples from the distribution of possible renderings consistent with the input and, even in the presence of ambiguity, is capable of rendering diverse and plausible novel views. To achieve this, our method makes use of existing 2D diffusion backbones but, crucially, incorporates geometry priors in the form of a 3D feature volume. This latent feature field captures the distribution over possible scene representations and improves our method's ability to generate view-consistent novel renderings. In addition to generating novel views, our method has the ability to autoregressively synthesize 3D-consistent sequences. We demonstrate state-of-the-art results on synthetic renderings and room-scale scenes; we also show compelling results for challenging, real-world objects., Comment: Project page: https://nvlabs.github.io/genvs