Abstract

We introduce MoRAG, a novel multi-part fusion based retrieval-augmented generation strategy for text-based human motion generation. The method enhances motion diffusion models by leveraging additional knowledge obtained through an improved motion retrieval process. By effectively prompting large language models (LLMs), we address spelling errors and rephrasing issues in motion retrieval. Our approach utilizes a multi-part retrieval strategy to improve the generalizability of motion retrieval across the language space. We create diverse samples through the spatial composition of the retrieved motions. Furthermore, by utilizing low-level, part-specific motion information, we can construct motion samples for unseen text descriptions. Our experiments demonstrate that our framework can serve as a plug-and-play module, improving the performance of motion diffusion models. Code, pretrained models and sample videos will be made available at: https://motionrag.github.io/

Abstract

Moving Target Defense (MTD) has emerged as a proactive and dynamic framework to counteract evolving cyber threats. Traditional MTD approaches often rely on assumptions about the attacker’s knowledge and behavior. However, real-world scenarios are inherently more complex, with adaptive attackers and limited prior knowledge of their payoffs and intentions. This paper introduces a novel approach to MTD using a Markov Decision Process (MDP) model that does not rely on predefined attacker payoffs. Our framework integrates the attacker’s real-time responses into the defender’s MDP using a dynamic Bayesian Network. By employing a factored MDP model, we provide a comprehensive and realistic representation of the system and its dependencies. We also incorporate incremental updates to an attack response predictor as new data emerges, ensuring an adaptive and robust defense mechanism. Additionally, we consider the costs of switching configurations in MTD, integrating them into the reward structure to balance execution and defense costs. We first highlight the challenges of the problem through a theoretical negative result on regret. However, our empirical evaluations, conducted in network and web application settings, demonstrate the framework’s effectiveness in environments characterized by high uncertainty and dynamically changing attack surfaces.

Abstract

Text driven diffusion models have shown remarkable capabilities in editing images. However, when editing 3D scenes, existing works mostly rely on training a NeRF for 3D editing. Recent NeRF editing methods leverages edit operations by deploying 2D diffusion models and project these edits into 3D space. They require strong positional priors alongside text prompt to identify the edit location. These methods are operational on small 3D scenes and are more generalized to particular scene. They require training for each specific edit and cannot be exploited in real-time edits. To address these limitations, we propose a novel method, FreeEdit, to make edits in training free manner using mesh representations as a substitute for NeRF. Training-free methods are now a possibility because of the advances in foundation model’s space. We leverage these models to bring a training-free alternative and introduce solutions for insertion, replacement and deletion. We consider insertion, replacement and deletion as basic blocks for performing intricate edits with certain combinations of these operations. Given a text prompt and a 3D scene, our model is capable of identifying what object should be inserted/replaced or deleted and location where edit should be performed. We also introduce a novel algorithm as part of FreeEdit to find the optimal location on grounding object for placement. We evaluate our model by comparing it with baseline models on a wide range of scenes using quantitative and qualitative metrics and showcase the merits of our method with respect to others.

Abstract

Deep learning assisted diagnosis for assessing the severity of various respiratory infections using chest Computed Tomography (CT) scan images has gained much attention after the COVID-19 pandemic. Major tasks while building such models require an understanding of the characteristic features associated with the disease, patient-to-patient variations and changes associated with disease severity. In this work, an attention-based Convolutional Neural Network (CNN) model with customized bottleneck residual module (Attn-CNN) is proposed for classifying CT images into three classes: COVID-19, Normal, and other Pneumonia. The efficacy of the model is evaluated by carrying out various experiments, such as effect of class imbalance, impact of attention module, generalizability of the model and providing visualization of model’s prediction for the interpretability of results. Comparative performance evaluation with five state-of-the-art deep architectures such as MobileNet, EfficientNet-B7, Inceptionv3, ResNet-50 and VGG-16, and with published models such as COVIDNet-CT, COVNet, COVID-Net CT2, etc. is discussed.

Abstract

No Results Found

Abstract

No Results Found

Abstract

Monuments were built for thousands of years as the most enduring and well-known representations of past civilizations. It is crucial to preserve these because they are national symbols with unique cultural and historical significance. The deterioration of the structures happens as a result of weathering, fire, natural disasters like earthquakes, construction flaws, and numerous other factors. In general, the issue with repairing and reinforcing regular buildings is very different from the issue with restoring monument buildings. The goal of keeping a monumental structure in use may be viewed as secondary in importance and any event as a result of the effort made to fulfil the main task. In the case of monumental buildings, emphasis is placed on the preservation of their aesthetic and historical aspects. Therefore, the restoration strives to protect and highlight the monument’s or building’s aesthetic and historical aspects. Choosing the appropriate material for the restoration depends on how the original structure looks. New materials must be compatible with those already used in the building. The goal of the present study is to provide an overview of the materials used to restore the structural integrity of historic structures and monuments that have been damaged by seismic activity. All of these buildings need extra attention because of their specific historical or architectural value, or because they are significant because they are still standing examples of a former way of life.

Abstract

Masonry structures cover a very wide range of monumental works in space and time, covering almost 6,000 years. The structural assessment of the load-carrying capacity of such old structures depends on several parameters such as: (a) material strength; (b) texture of the structure; (c) geometrical form and dimensions; (d) existing loading and constraints; and (e) expected loadings at failure. Detailed calculations are also required for the structural evaluation using most of the above parameters firstly by considering the gravity loads and secondly by the combination of gravity load and lateral action in areas of high seismicity. However, at the same time, it is equally significant for the structural engineer, who is involved in structural restoration design, to know in depth qualitatively the basic forms and their structural behaviour of monumental buildings. The structural engineer in charge must understand from the beginning, qualitatively, the structural behaviour of a building under consideration to various basic actions so that one can make a preliminary diagnosis of the causes that are responsible for the existing damages (cracks, drifts, settlements, splitting etc.). In this way, one will be in a position to focus on the detailed research, in situ and laboratory, to the critical points of the building, and can conduct the analysis and design using suitable models for a credible safety evaluation. Therefore, an effort is made in the present study to give a qualitative approach to the structural behaviour of the basic structural forms of masonry monumental and historical buildings, where the consequences of the seismic action are traumatic now and then. This article gives an outline and creates background information for further structural evaluation of the monuments.

Abstract

This paper considers the contextual multi-armed bandit (CMAB) problem with fairness and privacy guarantees in a federated environment. We consider merit-based exposure as the desired fair outcome, which provides exposure to each action in proportion to the reward associated. We model the algorithm's effectiveness using fairness regret, which captures the difference between fair optimal policy and the policy output by the algorithm. Applying fair CMAB algorithm to each agent individually leads to fairness regret linear in the number of agents. We propose that collaborative -- federated learning can be more effective and provide the algorithm Fed-FairX-LinUCB that also ensures differential privacy. The primary challenge in extending the existing privacy framework is designing the communication protocol for communicating required information across agents. A naive protocol can either lead to weaker privacy guarantees or higher regret. We design a novel communication protocol that allows for (i) Sub-linear theoretical bounds on fairness regret for Fed-FairX-LinUCB and comparable bounds for the private counterpart, Priv-FairX-LinUCB (relative to single-agent learning), (ii) Effective use of privacy budget in Priv-FairX-LinUCB. We demonstrate the efficacy of our proposed algorithm with extensive simulations-based experiments. We show that both Fed-FairX-LinUCB and Priv-FairX-LinUCB achieve near-optimal fairness regret.

Abstract

Restless multi-armed bandits (RMABs) generalize the multi-armed bandits where each arm exhibits Markovian behavior and transitions according to their transition dynamics. Solutions to RMAB exist for both offline and online cases. However, they do not consider the distribution of pulls among the arms. Studies have shown that optimal policies lead to unfairness, where some arms are not exposed enough. Existing works in fairness in RMABs focus heavily on the offline case, which diminishes their application in real-world scenarios where the environment is largely unknown. In the online scenario, we propose the first fair RMAB framework, where each arm receives pulls in proportion to its merit. We define the merit of an arm as a function of its stationary reward distribution. We prove that our algorithm achieves sublinear fairness regret in the single pull case , with being the total number of episodes. Empirically, we show that our algorithm performs well in the multi-pull scenario as well.