Search

Image tokenizers map images to sequences of discrete tokens, and are a crucial component of autoregressive transformer-based image generation. The tokens are typically associated with spatial locations in the input image, arranged in raster scan order, which is not ideal for autoregressive modeling. In this paper, we propose to tokenize the image spectrum instead, obtained from a discrete wavelet transform (DWT), such that the sequence of tokens represents the image in a coarse-to-fine fashion. Our tokenizer brings several advantages: 1) it leverages that natural images are more compressible at high frequencies, 2) it can take and reconstruct images of different resolutions without retraining, 3) it improves the conditioning for next-token prediction -- instead of conditioning on a partial line-by-line reconstruction of the image, it takes a coarse reconstruction of the full image, 4) it enables partial decoding where the first few generated tokens can reconstruct a coarse version of the image, 5) it enables autoregressive models to be used for image upsampling. We evaluate the tokenizer reconstruction metrics as well as multiscale image generation, text-guided image upsampling and editing.

Latent variable generative models have emerged as powerful tools for generative tasks including image and video synthesis. These models are enabled by pretrained autoencoders that map high resolution data into a compressed lower dimensional latent space, where the generative models can subsequently be developed while requiring fewer computational resources. Despite their effectiveness, the direct application of latent variable models to higher dimensional domains such as videos continues to pose challenges for efficient training and inference. In this paper, we propose an autoencoder that projects volumetric data onto a four-plane factorized latent space that grows sublinearly with the input size, making it ideal for higher dimensional data like videos. The design of our factorized model supports straightforward adoption in a number of conditional generation tasks with latent diffusion models (LDMs), such as class-conditional generation, frame prediction, and video interpolation. Our results show that the proposed four-plane latent space retains a rich representation needed for high-fidelity reconstructions despite the heavy compression, while simultaneously enabling LDMs to operate with significant improvements in speed and memory.

The origins of Uranus and Neptune are not fully understood. Their inclined rotation axes -- obliquities -- suggest that they experienced giant impacts during their formation histories. Simulations modeling their accretion from giant impacts among ~5 Earth masses planetary embryos -- with roughly unity impactors' mass ratios -- have been able to broadly match their current masses, final mass ratio, and obliquity. However, due to angular momentum conservation, planets produced in these impacts tend to rotate too fast, compared to Uranus and Neptune. One potential solution for this problem consists of invoking instead collisions of objects with large mass ratios (e.g. a proto-Uranus with 13 Mearth and an embryo of 1 Mearth). Smooth-particle hydrodynamics simulations show that in this scenario final planets tend to have rotation periods more consistent with those of Uranus and Neptune. Here we performed a large suite of N-body numerical simulations modelling the formation of Uranus and Neptune to compare these different dynamical views. Our simulations start with a population of protoplanets and account for the effects of type-I migration, inclination and eccentricity tidal damping. Our results show that although scenarios allowing for large impactors' mass ratio favour slower rotating planets, the probability of occurring collisions in these specific simulations is significantly low. This is because gas tidal damping is relatively less efficient for low-mass embryos (<~1 Merath) and, consequently, such objects are mostly scattered by more massive objects (~13 Mearth) instead of colliding with them. Altogether, our results show that the probability of broadly matching the masses, mass ratio, and rotation periods of Uranus and Neptune in these two competing formation scenarios is broadly similar, within a factor of ~2, with overall probabilities of the order of ~0.1-1%., Comment: 15 pages, 5 figures. Accepted for publication in Icarus

Similarities in the non-mass dependent isotopic composition of refractory elements with the bulk silicate Earth suggest that both the Earth and the Moon formed from the same material reservoir. On the other hand, the Moon's volatile depletion and isotopic composition of moderately volatile elements points to a global devolatilization processes, most likely during a magma ocean phase of the Moon. Here, we investigate the devolatilisation of the molten Moon due to a tidally-assisted hydrodynamic escape with a focus on the dynamics of the evaporated gas. Unlike the 1D steady-state approach of Charnoz et al. (2021), we use 2D time-dependent hydrodynamic simulations carried out with the FARGOCA code modified to take into account the magma ocean as a gas source. Near the Earth's Roche limit, where the proto-Moon likely formed, evaporated gases from the lunar magma ocean form a circum-Earth disk of volatiles, with less than 30% of material being re-accreted by the Moon. We find that the measured depletion of K and Na on the Moon can be achieved if the lunar magma-ocean had a surface temperature of about 1800-2000 K. After about 1000 years, a thermal boundary layer or a flotation crust forms a lid that inhibits volatile escape. Mapping the volatile velocity field reveals varying trends in the longitudes of volatile reaccretion on the Moon's surface: material is predominantly re-accreted on the trailing side when the Moon-Earth distance exceeds 3.5 Earth radii, suggesting a dichotomy in volatile abundances between the leading and trailing sides of the Moon. This dichotomy may provide insights on the tidal conditions of the early molten Earth. In conclusion, tidally-driven atmospheric escape effectively devolatilizes the Moon, matching the measured abundances of Na and K on timescales compatible with the formation of a thermal boundary layer or an anorthite flotation crust., Comment: Accepted for publication in Earth and Planetary Science Letters

Metal organic frameworks (MOFs) are nanoporous materials with high surface-to-volume ratio that have potential applications as gas sorbents. Sample quality is, however, often compromised and it is unclear how defects and surface contaminants affect the spectral properties of single MOF crystals. Raman micro-spectroscopy is a powerful tool for characterizing MOFs, yet spatial spectral heterogeneity distributions of single MOF crystals have not been reported so far. In this work, we use Raman micro-spectroscopy to characterize spatially isolated, single crystals of the MOF species HKUST-1. In a first step, we validate HKUST-1's Raman spectrum based on DFT simulations and we identify a previously unreported vibrational feature. In a second step, we acquire diffraction-limited, mode-selective Raman images of a single HKUST-1 crystals that reveal how the spectral variations are distributed across the crystal surface. In a third step, we statistically analyze the measured spectral peak positions and line widths for quantifying the variability occurring within the same crystal as well as between different crystals taken from the same batch. Finally, we explore how multivariate data analysis can aid feature identification in Raman images of single MOF crystals. For enabling validation and reuse, we have made the spectroscopic data and simulation code publicly available.

The User-Managed Access (UMA) extension to OAuth 2.0 is a promising candidate for increasing Digital Trust in personal data ecosystems like Solid. With minor modifications, it can achieve many requirements regarding usage control and transaction contextualization, even though additional specification is needed to address delegation of control and retraction of usage policies.

A level graph is the data of a pair $(G,\pi)$ consisting of a finite graph $G$ and an ordered partition $\pi$ on the set of vertices of $G$. To each level graph on $n$ vertices we associate a polytope in $\mathbb R^n$ called its residue polytope. We show that residue polytopes are compatible with each other in the sense that if $\pi'$ is a coarsening of $\pi$, then the polytope associated to $(G,\pi)$ is a face of the one associated to $(G,\pi')$. Moreover, they form all the faces of the residue polytope of $G$, defined as the polytope associated to the level graph with the trivial ordered partition. The results are used in a companion work to describe limits of spaces of Abelian differentials on families of Riemann surfaces approaching a stable Riemann surface on the boundary of the moduli space., Comment: 18 pages, 3 figures

We describe the limits of canonical series along families of curves degenerating to a nodal curve which is general for its topology, in the weak sense that the branches over nodes on each of its components are in general position. We define a fan structure on the space of edge lengths on the dual graph of the limit curve, and construct a projective variety parametrizing the limits, organized in strata associated to the cones of this fan., Comment: 79 pages, 10 figures, 3 appendices

The pepenaactivity is part of the informal sector, characterized by precariousness and social invisibility; however, it is a key element in the recycling production chain, providing an economic income for many people. In the city of Chihuahua, this activity is carried out on a significant scale. Therefore, this study analyzes and dignifies the recycling of urban solid waste from the perspective of scavengers, also explaining their relationship with other social actors at the city's final disposal site. The study starts from the premise that waste picking is, in the current era of global environmental crisis, an effective way to conserve resources and reduce environmental impacts. Methodologically, participatory action research was used as a strategy to raise awareness among scavengers about the importance of their work and to present their community organization as an example to other urban waste picking groups in different locations. Finally, we found issues in these groups such as informality, lack of legal support,specific health risks, and, more broadly, the absence of public policies to recognize this activity as valuable in addressing the global environmental crisis, Comment: in Spanish language

A long-standing challenge is the design of chips resilient to faults and glitches. Both fine-grained gate diversity and coarse-grained modular redundancy have been used in the past. However, these approaches have not been well-studied under other threat models where some stakeholders in the supply chain are untrusted. Increasing digital sovereignty tensions raise concerns regarding the use of foreign off-the-shelf tools and IPs, or off-sourcing fabrication, driving research into the design of resilient chips under this threat model. This paper addresses a threat model considering three pertinent attacks to resilience: distribution, zonal, and compound attacks. To mitigate these attacks, we introduce the \texttt{ResiLogic} framework that exploits \textit{Diversity by Composability}: constructing diverse circuits composed of smaller diverse ones by design. This gives designer the capability to create circuits at design time without requiring extra redundancy in space or cost. Using this approach at different levels of granularity is shown to improve the resilience of circuit design in \texttt{ResiLogic} against the three considered attacks by a factor of five. Additionally, we also make a case to show how E-Graphs can be utilized to generate diverse circuits under given rewrite rules.

Curvilinear structures are present in various fields in image processing such as blood vessels in medical imaging or roads in remote sensing. Their detection is crucial for many applications. In this article, we propose an unsupervised plug-and-play framework for the segmentation of curvilinear structures that focuses on the preservation of their connectivity. This framework includes an algorithm for generating realistic pairs of connected/disconnected curvilinear structures and a reconnecting regularization operator that can be learned from a synthetic dataset. Once learned, this regularization operator can be plugged into a variational segmentation scheme and used to segment curvilinear structure images without requiring annotations. We demonstrate the interest of our approach on the segmentation of vascular images both in 2D and 3D and compare its results with classic unsupervised and deep learning-based approach. Comparative evaluations against unsupervised classic and deep learning-based methods highlight the superior performance of our approach, showcasing remarkable improvements in preserving the connectivity of curvilinear structures (approximately 90% in 2D and 70% in 3D). We finally showcase the good generalizability behavior of our approach on two different applications : road cracks and porcine corneal cells segmentations.

This paper presents a sociotechnical vision for managing personal data, including cookies, within Web browsers. We first present our vision for a future of semi-automated data governance on the Web, using policy languages to describe data terms of use, and having browsers act on behalf of users to enact policy-based controls. Then, we present an overview of the technical research required to {prove} that existing policy languages express a sufficient range of concepts for describing cookie policies on the Web today. We view this work as a stepping stone towards a future of semi-automated data governance at Web-scale, which in the long term will also be used by next-generation Web technologies such as Web agents and Solid., Comment: Submitted to "NeXt-generation Data Governance workshop 2024"; available on OpenReview at https://openreview.net/forum?id=Bhia6mPaCF

In this paper, we propose Sparse High Rank Adapters (SHiRA) that directly finetune 1-2% of the base model weights while leaving others unchanged, thus, resulting in a highly sparse adapter. This high sparsity incurs no inference overhead, enables rapid switching directly in the fused mode, and significantly reduces concept-loss during multi-adapter fusion. Our extensive experiments on LVMs and LLMs demonstrate that finetuning merely 1-2% parameters in the base model is sufficient for many adapter tasks and significantly outperforms Low Rank Adaptation (LoRA). We also show that SHiRA is orthogonal to advanced LoRA methods such as DoRA and can be easily combined with existing techniques., Comment: Published at ICML 2024 Workshop on Foundation Models in the Wild. arXiv admin note: substantial text overlap with arXiv:2406.13175

Predicting the geometrical evolution of the pore space in geological formations due to fluid-solid interactions has applications in reservoir engineering, oil recovery, and geological storage of carbon dioxide. However, modeling frameworks that combine fluid flow with physical and chemical processes at a rock's pore scale are scarce. Here, we report a method for modeling a rock's pore space as a network of connected capillaries and to simulate the capillary diameter modifications caused by reactive flow processes. Specifically, we model mineral erosion, deposition, dissolution, and precipitation processes by solving the transport equations iteratively, computing diameter changes within each capillary of the network simultaneously. Our automated modeling framework enables simulations on digital rock samples as large as (1.125mm)$^3$ with 125$\times 10^6$ voxels within seconds of CPU time per iteration. As an application of the computational method, we have simulated brine injection and calcium carbonate precipitation in sandstone. For quantitatively comparing simulation results obtained with models predicting either a constant or a flow-rate dependent precipitation, we track the time-dependent capillary diameter distribution as well as the permeability of the connected pore space. For validation and reuse, we have made the automated simulation workflow, the reactive flow model library, and the digital rock samples available in public repositories., Comment: Main article: 12 pages, 1 equation, 2 tables and 4 figures. Supplementary Information: 6 pages, 11 equations and 3 figures. Includes DOI for accompanying digital rock data, URL of github repository for flow simulator code, and of accompanying processing python code for automation of the scientific workflows

Computational offload to hardware accelerators is gaining traction due to increasing computational demands and efficiency challenges. Programmable hardware, like FPGAs, offers a promising platform in rapidly evolving application areas, with the benefits of hardware acceleration and software programmability. Unfortunately, such systems composed of multiple hardware components must consider integrity in the case of malicious components. In this work, we propose Samsara, the first secure and resilient platform that derives, from Byzantine Fault Tolerant (BFT), protocols to enhance the computing resilience of programmable hardware. Samsara uses a novel lightweight hardware-based BFT protocol for Systems-on-Chip, called H-Quorum, that implements the theoretical-minimum latency between applications and replicated compute nodes. To withstand malicious behaviors, Samsara supports hardware rejuvenation, which is used to replace, relocate, or diversify faulty compute nodes. Samsara's architecture ensures the security of the entire workflow while keeping the latency overhead, of both computation and rejuvenation, close to the non-replicated counterpart., Comment: To be published in The 43rd International Symposium on Reliable Distributed Systems (SRDS 2024)

Low Rank Adaptation (LoRA) has gained massive attention in the recent generative AI research. One of the main advantages of LoRA is its ability to be fused with pretrained models adding no overhead during inference. However, from a mobile deployment standpoint, we can either avoid inference overhead in the fused mode but lose the ability to switch adapters rapidly, or suffer significant (up to 30% higher) inference latency while enabling rapid switching in the unfused mode. LoRA also exhibits concept-loss when multiple adapters are used concurrently. In this paper, we propose Sparse High Rank Adapters (SHiRA), a new paradigm which incurs no inference overhead, enables rapid switching, and significantly reduces concept-loss. Specifically, SHiRA can be trained by directly tuning only 1-2% of the base model weights while leaving others unchanged. This results in a highly sparse adapter which can be switched directly in the fused mode. We further provide theoretical and empirical insights on how high sparsity in SHiRA can aid multi-adapter fusion by reducing concept loss. Our extensive experiments on LVMs and LLMs demonstrate that finetuning only a small fraction of the parameters in the base model is sufficient for many tasks while enabling both rapid switching and multi-adapter fusion. Finally, we provide a latency- and memory-efficient SHiRA implementation based on Parameter-Efficient Finetuning (PEFT) Library. This implementation trains at nearly the same speed as LoRA while consuming lower peak GPU memory, thus making SHiRA easy to adopt for practical use cases.

While Low-Rank Adaptation (LoRA) has proven beneficial for efficiently fine-tuning large models, LoRA fine-tuned text-to-image diffusion models lack diversity in the generated images, as the model tends to copy data from the observed training samples. This effect becomes more pronounced at higher values of adapter strength and for adapters with higher ranks which are fine-tuned on smaller datasets. To address these challenges, we present FouRA, a novel low-rank method that learns projections in the Fourier domain along with learning a flexible input-dependent adapter rank selection strategy. Through extensive experiments and analysis, we show that FouRA successfully solves the problems related to data copying and distribution collapse while significantly improving the generated image quality. We demonstrate that FouRA enhances the generalization of fine-tuned models thanks to its adaptive rank selection. We further show that the learned projections in the frequency domain are decorrelated and prove effective when merging multiple adapters. While FouRA is motivated for vision tasks, we also demonstrate its merits for language tasks on the GLUE benchmark.

We show that tropicalization of linear series on curves gives rise to two-parameter families of tilings by polymatroids, with one parameter arising from the theory of divisors on tropical curves and the other from the reduction of linear series of rational functions in non-Archimedean geometry. In order to do this, we introduce a general framework that produces tilings of vector spaces and their subsets by polymatroids. We furthermore show that these tilings are regular and relate them to work by Kapranov and Lafforgue on Chow quotients of Grassmannians., Comment: 55 pages, 7 figures. Comments are welcome

The Data Privacy Vocabulary (DPV), developed by the W3C Data Privacy Vocabularies and Controls Community Group (DPVCG), enables the creation of machine-readable, interoperable, and standards-based representations for describing the processing of personal data. The group has also published extensions to the DPV to describe specific applications to support legislative requirements such as the EU's GDPR. The DPV fills a crucial niche in the state of the art by providing a vocabulary that can be embedded and used alongside other existing standards such as W3C ODRL, and which can be customised and extended for adapting to specifics of use-cases or domains. This article describes the version 2 iteration of the DPV in terms of its contents, methodology, current adoptions and uses, and future potential. It also describes the relevance and role of DPV in acting as a common vocabulary to support various regulatory (e.g. EU's DGA and AI Act) and community initiatives (e.g. Solid) emerging across the globe., Comment: Accepted for Presentation in The 23rd International Semantic Web Conference (ISWC 2024)

Accurate segmentation of vascular networks is essential for computer-aided tools designed to address cardiovascular diseases. Despite more than thirty years of research, it remains a challenge to obtain vascular segmentation results that preserve the connectivity of the underlying vascular network. Yet connectivity is one of the key feature of these tools. In this work, we propose a post-processing algorithm aiming to reconnect vascular structures that have been disconnected by a segmentation algorithm. Connectivity being a complex property to model explicity, we propose to learn this geometric feature either through synthetic data or annotations of the application of interest. The resulting post-processing model can be used on the output of any supervised or unsupervised vascular segmentation algorithm. We show that this post-processing effectively restores the connectivity of vascular networks both in 2D and 3D images, leading to improved overall segmentation results.

Amidst the surge in deep learning-based password guessing models, challenges of generating high-quality passwords and reducing duplicate passwords persist. To address these challenges, we present PagPassGPT, a password guessing model constructed on Generative Pretrained Transformer (GPT). It can perform pattern guided guessing by incorporating pattern structure information as background knowledge, resulting in a significant increase in the hit rate. Furthermore, we propose D&C-GEN to reduce the repeat rate of generated passwords, which adopts the concept of a divide-and-conquer approach. The primary task of guessing passwords is recursively divided into non-overlapping subtasks. Each subtask inherits the knowledge from the parent task and predicts succeeding tokens. In comparison to the state-of-the-art model, our proposed scheme exhibits the capability to correctly guess 12% more passwords while producing 25% fewer duplicates., Comment: Be accepted by DSN 2024

With recent legislation on the right to be forgotten, machine unlearning has emerged as a crucial research area. It facilitates the removal of a user's data from federated trained machine learning models without the necessity for retraining from scratch. However, current machine unlearning algorithms are confronted with challenges of efficiency and validity. To address the above issues, we propose a new framework, named Goldfish. It comprises four modules: basic model, loss function, optimization, and extension. To address the challenge of low validity in existing machine unlearning algorithms, we propose a novel loss function. It takes into account the loss arising from the discrepancy between predictions and actual labels in the remaining dataset. Simultaneously, it takes into consideration the bias of predicted results on the removed dataset. Moreover, it accounts for the confidence level of predicted results. Additionally, to enhance efficiency, we adopt knowledge a distillation technique in the basic model and introduce an optimization module that encompasses the early termination mechanism guided by empirical risk and the data partition mechanism. Furthermore, to bolster the robustness of the aggregated model, we propose an extension module that incorporates a mechanism using adaptive distillation temperature to address the heterogeneity of user local data and a mechanism using adaptive weight to handle the variety in the quality of uploaded models. Finally, we conduct comprehensive experiments to illustrate the effectiveness of proposed approach.

Automated molecular simulations are used extensively for predicting material properties. Typically, these simulations exhibit two regimes: a dynamic equilibration part, followed by a steady state. For extracting observable properties, the simulations must first reach a steady state so that thermodynamic averages can be taken. However, as equilibration depends on simulation conditions, predicting the optimal number of simulation steps a priori is impossible. Here, we demonstrate the application of the Marginal Standard Error Rule (MSER) for automatically identifying the optimal truncation point in Grand Canonical Monte Carlo (GCMC) simulations. This novel automatic procedure determines the point in which steady state is reached, ensuring that figures-of-merits are extracted in an objective, accurate, and reproducible fashion. In the case of GCMC simulations of gas adsorption in metal-organic frameworks, we find that this methodology reduces the computational cost by up to 90%. As MSER statistics are independent of the simulation method that creates the data, this library is, in principle, applicable to any time series analysis in which equilibration truncation is required. The open-source Python implementation of our method, pyMSER, is publicly available for reuse and validation at https://github.com/IBM/pymser., Comment: 22 pages, 19 figures

We contribute NeuralSolver, a novel recurrent solver that can efficiently and consistently extrapolate, i.e., learn algorithms from smaller problems (in terms of observation size) and execute those algorithms in large problems. Contrary to previous recurrent solvers, NeuralSolver can be naturally applied in both same-size problems, where the input and output sizes are the same, and in different-size problems, where the size of the input and output differ. To allow for this versatility, we design NeuralSolver with three main components: a recurrent module, that iteratively processes input information at different scales, a processing module, responsible for aggregating the previously processed information, and a curriculum-based training scheme, that improves the extrapolation performance of the method. To evaluate our method we introduce a set of novel different-size tasks and we show that NeuralSolver consistently outperforms the prior state-of-the-art recurrent solvers in extrapolating to larger problems, considering smaller training problems and requiring less parameters than other approaches.

The increasing interest in Autonomous Vehicles (AV) is notable due to business, safety, and performance reasons. While there is salient success in recent AV architectures, hinging on the advancements in AI models, there is a growing number of fatal incidents that impedes full AVs from going mainstream. This calls for the need to revisit the fundamentals of building safety-critical AV architectures. However, this direction should not deter leveraging the power of AI. To this end, we propose Savvy, a new trustworthy intelligent AV architecture that achieves the best of both worlds. Savvy makes a clear separation between the control plane and the data plane to guarantee the safety-first principles. The former assume control to ensure safety using design-time defined rules, while launching the latter for optimizing decisions as much as possible within safety time-bounds. This is achieved through guided Time-aware predictive quality degradation (TPQD): using dynamic ML models that can be tuned to provide either richer or faster outputs based on the available safety time bounds. For instance, Savvy allows to safely identify an elephant as an obstacle (a mere object) the earliest possible, rather than optimally recognizing it as an elephant when it is too late. This position paper presents the Savvy's motivations and concept, whereas empirical evaluation is a work in progress.

Cosmological analyses using galaxy clusters in optical/NIR photometric surveys require robust characterization of their galaxy content. Precisely determining which galaxies belong to a cluster is crucial. In this paper, we present the COlor Probabilistic Assignment of Clusters And BAyesiaN Analysis (Copacabana) algorithm. Copacabana computes membership probabilities for {\it all} galaxies within an aperture centred on the cluster using photometric redshifts, colours, and projected radial probability density functions. We use simulations to validate Copacabana and we show that it achieves up to 89\% membership accuracy with a mild dependency on photometric redshift uncertainties and choice of aperture size. We find that the precision of the photometric redshifts has the largest impact on the determination of the membership probabilities followed by the choice of the cluster aperture size. We also quantify how much these uncertainties in the membership probabilities affect the stellar mass--cluster mass scaling relation, a relation that directly impacts cosmology. Using the sum of the stellar masses weighted by membership probabilities ($\mu_{\star}$) as the observable, we find that Copacabana can reach an accuracy of 0.06 dex in the measurement of the scaling relation. These results indicate the potential of Copacabana and $\mu_{\star}$ to be used in cosmological analyses of optically selected clusters in the future., Comment: 14 pages, 10 figures, submitted to MNRAS

Carbon dioxide (CO2) trapping in capillary networks of reservoir rocks is a pathway to long-term geological storage. At pore scale, the CO2 trapping potential depends on injection pressure, temperature, and the rock's interaction with the surrounding fluids. Modeling this interaction requires adequate representations of both capillary volume and surface. For the lack of scalable representations, however, the prediction of a rock's CO2 storage potential has been challenging. Here, we report how to represent a rock's pore space by statistically sampled capillary networks (ssCN) that preserve morphological rock characteristics. We have used the ssCN method to simulate CO2 drainage within a representative sandstone sample at reservoir pressures and temperatures, exploring intermediate- and CO2-wet conditions. This wetting regime is often neglected, despite evidence of plausibility. By raising pressure and temperature we observe increasing CO2 penetration within the capillary network. For contact angles approaching 90 degrees, the CO2 saturation exhibits a pronounced maximum reaching 80 percent of the accessible pore volume. This is about twice as high as the saturation values reported previously. For enabling validation of our results and a broader application of our methodology, we have made available the rock tomography data, the digital rock computational workflows, and the ssCN models used in this study., Comment: Main article: 16 pages and 3 figures. Supplementary Information: 17 pages, 3 equations, 13 figures and 3 tables. Includes DOI for accompanying digital rock data, URL of github repository for flow simulator code, and of accompanying digital rock processing python code and automation of the scientific workflows

We introduce a framework for intrinsic latent diffusion models operating directly on the surfaces of 3D shapes, with the goal of synthesizing high-quality textures. Our approach is underpinned by two contributions: field latents, a latent representation encoding textures as discrete vector fields on the mesh vertices, and field latent diffusion models, which learn to denoise a diffusion process in the learned latent space on the surface. We consider a single-textured-mesh paradigm, where our models are trained to generate variations of a given texture on a mesh. We show the synthesized textures are of superior fidelity compared those from existing single-textured-mesh generative models. Our models can also be adapted for user-controlled editing tasks such as inpainting and label-guided generation. The efficacy of our approach is due in part to the equivariance of our proposed framework under isometries, allowing our models to seamlessly reproduce details across locally similar regions and opening the door to a notion of generative texture transfer., Comment: CVPR 2024. Code and additional visualizations available: https://single-mesh-diffusion.github.io/

A recent paper by Gupta et al. (EuroSys'23) challenged the usefulness of trusted component (TC) based Byzantine fault-tolerant (BFT) protocols to lower the replica group size from $3f+1$ to $2f+1$, identifying three limitations of such protocols and proposing that TCs should be used instead to improve the performance of BFT protocols. Here, we point out flaws in both arguments and advocate that the most worthwhile use of TCs in BFT protocols is indeed to make them as resilient as crash fault-tolerant (CFT) protocols, which can tolerate up to $f$ faulty replicas using $2f+1$ replicas.