Related papers: SWAN: World-Aware Adaptive Multimodal Networks for…

Predicting multiple real-world tasks in a single model often requires a particularly diverse feature space. Multimodal (MM) models aim to extract the synergistic predictive potential of multiple data types to create a shared feature space…

We propose a new network architecture, Gated Attention Networks (GaAN), for learning on graphs. Unlike the traditional multi-head attention mechanism, which equally consumes all attention heads, GaAN uses a convolutional sub-network to…

Vision Transformers have achieved remarkable progresses, among which Swin Transformer has demonstrated the tremendous potential of Transformer for vision tasks. It surmounts the key challenge of high computational complexity by performing…

Geo-distributed analytics (GDA) frameworks transfer large datasets over the wide-area network (WAN). Yet existing frameworks often ignore the WAN topology. This disconnect between WAN-bound applications and the WAN itself results in missed…

Deep neural networks (DNNs) can enable precise control while maintaining low computational costs by circumventing the need for dynamic modeling. However, the deployment of such black-box approaches remains challenging for heavy-duty wheeled…

In this paper, we propose a destination-aware adaptive traffic flow rule aggregation (DATA) mechanism for facilitating traffic flow monitoring in SDN-based networks. This method adapts the number of flow table entries in SDN switches…

Robust multimodal systems must remain effective when some modalities are noisy, degraded, or unreliable. Existing multimodal fusion methods often learn modality selection jointly with representation learning, making it difficult to…

Self-attention networks (SANs) with selective mechanism has produced substantial improvements in various NLP tasks by concentrating on a subset of input words. However, the underlying reasons for their strong performance have not been well…

This paper presents a novel generative framework for learning shared latent representations across multimodal data. Many advanced multimodal methods focus on capturing all combinations of modality-specific details across inputs, which can…

Autonomous systems and smart-industry deployments increasingly split computation across near-sensor, edge, and cloud resources, where tight energy, latency, and reliability budgets demand run-time adaptivity. In practice, deciding what to…

An emerging paradigm in vision-and-language navigation (VLN) is the use of history-aware multi-modal transformer models. Given a language instruction, these models process observation and navigation history to predict the most appropriate…

Depth prediction is a critical problem in robotics applications especially autonomous driving. Generally, depth prediction based on binocular stereo matching and fusion of monocular image and laser point cloud are two mainstream methods.…

Traffic congestion in urban road networks leads to longer trip times and higher emissions, especially during peak periods. While the Shortest Path First (SPF) algorithm is optimal for a single vehicle in a static network, it performs poorly…

Neural networks are increasingly used as fast surrogate models across various domains, but unconstrained predictions can violate physical, operational, or safety requirements. We propose SnareNet, a feasibility-controlled architecture to…

The application of unsupervised domain adaptation (UDA)-based fault diagnosis methods has shown significant efficacy in industrial settings, facilitating the transfer of operational experience and fault signatures between different…

The recent advances in deep neural networks have convincingly demonstrated high capability in learning vision models on large datasets. Nevertheless, collecting expert labeled datasets especially with pixel-level annotations is an extremely…

The recent advances in deep transfer learning reveal that adversarial learning can be embedded into deep networks to learn more transferable features to reduce the distribution discrepancy between two domains. Existing adversarial domain…

Simulation-to-real domain adaptation for semantic segmentation has been actively studied for various applications such as autonomous driving. Existing methods mainly focus on a single-source setting, which cannot easily handle a more…

LPWANs have become ubiquitous due to their ability to connect sensors over large geographic areas in a single hop. It is, however, very challenging to achieve massive scalability in LPWANs, where numerous sensors can transmit data…

Unsupervised Domain Adaptation (UDA) aims to utilize labeled data from a source domain to solve tasks in an unlabeled target domain, often hindered by significant domain gaps. Traditional CNN-based methods struggle to fully capture complex…