Green Energy

A Rochester team engineered a new type of solar thermoelectric generator that produces 15 times more power than earlier versions. By enhancing heat absorption and dissipation rather than tweaking semiconductor materials, they dramatically improved efficiency and demonstrated practical applications like powering LEDs.

Danish wind farm developer Orsted says the Revolution Wind project is about 80% complete, with 45 out of its 65 turbines already installed. The post Trump halts work on New England offshore wind project that’s nearly complete appeared first on Boston.com.

R.J. Scaringe told the Plugged-In Podcast that he sees the promise of LFP and is stilll skeptical of solid-state technology.

Renewable power executives say difficulty getting permits, rising costs due to tariffs and the end of key tax credits are making it tough to plan.

Thirty-four teams from all over the world are competing to win the 2025 Bridgestone World Solar Challenge.

NASA's Mars Reconnaissance Orbiter recently photographed Candor Chasma in Mars' Valles Marineris, the largest canyon in the entire solar system.

Researchers at the University of British Columbia have shown that a small bench-top reactor can enhance nuclear fusion rates by electrochemically loading a metal with deuterium fuel. Unlike massive magnetic confinement reactors, their experiment uses a room-temperature setup that packs deuterium into palladium like a sponge, boosting the likelihood of fusion events.

The order to stop construction on Revolution Wind off the coast of Rhode Island is part of a campaign against renewable energy.

The energy transition must also focus on small, behind-the-meter community and household systems. The model needs expanding, including peer-to-peer trading.  The post Community energy zones can repair broken energy market, and offer fair value for solar, EVs and batteries appeared first on RenewEconomy.

The guidance aims to provide communities and first responders with best practices for safe BESS installation, operation and emergency response. The post EPA issues battery storage safety guidelines appeared first on Power Engineering.

Battery-swapping Nio has just upgraded its six-seater flagship. It's a lot nicer, while being significantly cheaper, too.

Romania has awarded power deals to more than 1.2GW of onshore wind capacity in its latest contracts for difference (CfD) auction.

Battery production is outpacing demand, which means prices should continue to fall.

Despite this, Jeep says the all-new Cherokee, which is hybrid-only for now, can drive at up to 62 mph on electric power.

Open source Surya system promises early alerts for space weather that can fry satellites and grids Boffins at IBM and NASA have concocted an AI model to help predict the weather, but this time it is taking on space weather that might disrupt satellites and spacecraft, possibly even terrestrial power grids and the internet.…

Colombia is offering 15-year power deals in South America's first offtake auction for offshore wind.

arXiv:2508.15537v1 Announce Type: new Abstract: Extracting narrow roads from high-resolution remote sensing imagery remains a significant challenge due to their limited width, fragmented topology, and frequent occlusions. To address these issues, we propose D3FNet, a Dilated Dual-Stream Differential Attention Fusion Network designed for fine-grained road structure segmentation in remote perception systems. Built upon the encoder-decoder backbone of D-LinkNet, D3FNet introduces three key innovations:(1) a Differential Attention Dilation Extraction (DADE) module that enhances subtle road features while suppressing background noise at the bottleneck; (2) a Dual-stream Decoding Fusion Mechanism (DDFM) that integrates original and attention-modulated features to balance spatial precision with semantic context; and (3) a multi-scale dilation strategy (rates 1, 3, 5, 9) that mitigates gridding artifacts and improves continuity in narrow road prediction. Unlike conventional models that

arXiv:2508.15517v1 Announce Type: new Abstract: Range and performance are key customer-relevant properties of electric vehicles. Both degrade over time due to battery aging, thus impacting business decisions throughout a vehicle's lifecycle, such as efficient utilization and asset valuation. For practical assessment, aging is often simplified into a single figure of merit - the state of health - typically defined by the battery pack's remaining capacity or energy. However, no standardized method for measuring the state of health at the vehicle level has been established, leaving both academia and industry without a clear consensus. Ultimately, standardization is crucial to increase transparency and build confidence in the long-term reliability of electric vehicles' battery packs. In this article, we propose a standard measurement procedure for assessing the capacity- and energy-based state of health, leveraging onboard charging to enable reproducibility and scalability. Additionally,

arXiv:2508.15505v1 Announce Type: new Abstract: Multimodal Image Fusion (MMIF) aims to integrate complementary information from different imaging modalities to overcome the limitations of individual sensors. It enhances image quality and facilitates downstream applications such as remote sensing, medical diagnostics, and robotics. Despite significant advancements, current MMIF methods still face challenges such as modality misalignment, high-frequency detail destruction, and task-specific limitations. To address these challenges, we propose AdaSFFuse, a novel framework for task-generalized MMIF through adaptive cross-domain co-fusion learning. AdaSFFuse introduces two key innovations: the Adaptive Approximate Wavelet Transform (AdaWAT) for frequency decoupling, and the Spatial-Frequency Mamba Blocks for efficient multimodal fusion. AdaWAT adaptively separates the high- and low-frequency components of multimodal images from different scenes, enabling fine-grained extraction and

arXiv:2508.15476v1 Announce Type: new Abstract: Medical image segmentation plays a pivotal role in disease diagnosis and treatment planning, particularly in resource-constrained clinical settings where lightweight and generalizable models are urgently needed. However, existing lightweight models often compromise performance for efficiency and rarely adopt computationally expensive attention mechanisms, severely restricting their global contextual perception capabilities. Additionally, current architectures neglect the channel redundancy issue under the same convolutional kernels in medical imaging, which hinders effective feature extraction. To address these challenges, we propose LGMSNet, a novel lightweight framework based on local and global dual multiscale that achieves state-of-the-art performance with minimal computational overhead. LGMSNet employs heterogeneous intra-layer kernels to extract local high-frequency information while mitigating channel redundancy. In addition, the

arXiv:2508.15312v1 Announce Type: new Abstract: Predicting user influence in social networks is a critical problem, and hypergraphs, as a prevalent higher-order modeling approach, provide new perspectives for this task. However, the absence of explicit cascade or infection probability data makes it particularly challenging to infer influence in hypergraphs. To address this, we introduce HIP, a unified and model-independent framework for influence prediction without knowing the underlying spreading model. HIP fuses multi-dimensional centrality indicators with a temporally reinterpreted distance matrix to effectively represent node-level diffusion capacity in the absence of observable spreading. These representations are further processed through a multi-hop Hypergraph Neural Network (HNN) to capture complex higher-order structural dependencies, while temporal correlations are modeled using a hybrid module that combines Long Short-Term Memory (LSTM) networks and Neural Ordinary

arXiv:2508.15175v1 Announce Type: new Abstract: This paper focuses on the privacy-preserving multi-sensor fusion estimation (MSFE) problem with differential privacy considerations. Most existing research efforts are directed towards the exploration of traditional differential privacy, also referred to as centralized differential privacy (CDP). It is important to note that CDP is tailored to protect the privacy of statistical data at fusion center such as averages and sums rather than individual data at sensors, which renders it inappropriate for MSFE. Additionally, the definitions and assumptions of CDP are primarily applicable for large-scale systems that require statistical results mentioned above. Therefore, to address these limitations, this paper introduces a more recent advancement known as \emph{local differential privacy (LDP)} to enhance the privacy of MSFE. We provide some rigorous definitions about LDP based on the intrinsic properties of MSFE rather than directly

Astronomers using the NASA/ESA/CSA James Webb Space Telescope have found strong evidence of a giant planet orbiting a

Australian households are not just putting in more solar batteries, they are putting in bigger ones. And that has big implications for how the grid is managed. The post Stunning success of home battery boom will slash rooftop solar exports, and reshape the grid – again appeared first on RenewEconomy.

Developers have already added 12 GW of utility scale solar in the U.S. so far in 2025, and with another 21 GW planned by the end of the year, solar is on track to account for half or more of all new generating capacity in the country this year. The post Half of new generation capacity in the U.S. will come from solar this year – report appeared first on Power Engineering.

Astronomers have picked up evidence of an Earth-sized world, distinct from the previously hypothesised Planet Nine and Planet X, that might be warping the orbits of objects beyond Neptune

A giant bubble of gas and dust surrounds the red supergiant DFK 52, likely created in a powerful outburst 4,000 years ago. Astronomers are baffled at how the star survived without going supernova, and suspect a hidden companion may have played a role. This discovery could reveal clues about the final stages of massive stars.

The new open-source AI model, Soraya, is trained on nine years of satellite imagery data and can accurately predict the sun's activity up to two hours into the future. It's 16% more effective than any other tool currently available.

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. On the ground in Ukraine’s largest Starlink repair shop Starlink is absolutely critical to Ukraine’s ability to continue in the fight against Russia. It’s how troops in battle zones stay connected with faraway…

Researchers said the breakthrough "paves the way for electronics powered by the ambient light already present in our lives."

The system integrates whole battery packs, including inverters, as a buffer in the factory’s energy supply.

arXiv:2508.14509v1 Announce Type: cross Abstract: This paper proposes a high-precision semantic segmentation method based on an improved TransUNet architecture to address the challenges of complex lesion structures, blurred boundaries, and significant scale variations in skin lesion images. The method integrates a transformer module into the traditional encoder-decoder framework to model global semantic information, while retaining a convolutional branch to preserve local texture and edge features. This enhances the model's ability to perceive fine-grained structures. A boundary-guided attention mechanism and multi-scale upsampling path are also designed to improve lesion boundary localization and segmentation consistency. To verify the effectiveness of the approach, a series of experiments were conducted, including comparative studies, hyperparameter sensitivity analysis, data augmentation effects, input resolution variation, and training data split ratio tests. Experimental results

arXiv:2508.14719v1 Announce Type: new Abstract: Photon-Counting Computed Tomography (PCCT) is a novel imaging modality that simultaneously acquires volumetric data at multiple X-ray energy levels, generating separate volumes that capture energy-dependent attenuation properties. Attenuation refers to the reduction in X-ray intensity as it passes through different tissues or materials. This spectral information enhances tissue and material differentiation, enabling more accurate diagnosis and analysis. However, the resulting multivolume datasets are often complex and redundant, making visualization and interpretation challenging. To address these challenges, we propose a method for fusing spectral PCCT data into a single representative volume that enables direct volume rendering and segmentation by leveraging both shared and complementary information across different channels. Our approach starts by computing 2D histograms between pairs of volumes to identify those that exhibit

arXiv:2508.14661v1 Announce Type: new Abstract: Aiming to enhance the consistency and thus long-term accuracy of Extended Kalman Filters for terrestrial vehicle localization, this paper introduces the Manifold Error State Extended Kalman Filter (M-ESEKF). By representing the robot's pose in a space with reduced dimensionality, the approach ensures feasible estimates on generic smooth surfaces, without introducing artificial constraints or simplifications that may degrade a filter's performance. The accompanying measurement models are compatible with common loosely- and tightly-coupled sensor modalities and also implicitly account for the ground geometry. We extend the formulation by introducing a novel correction scheme that embeds additional domain knowledge into the sensor data, giving more accurate uncertainty approximations and further enhancing filter consistency. The proposed estimator is seamlessly integrated into a validated modular state estimation framework, demonstrating

arXiv:2508.14597v1 Announce Type: new Abstract: Fire outbreaks pose critical threats to human life and infrastructure, necessitating high-fidelity early-warning systems that detect combustion precursors such as smoke. However, smoke plumes exhibit complex spatiotemporal dynamics influenced by illumination variability, flow kinematics, and environmental noise, undermining the reliability of traditional detectors. To address these challenges without the logistical complexity of multi-sensor arrays, we propose an information-fusion framework by integrating smoke feature representations extracted from monocular imagery. Specifically, a Two-Phase Uncertainty-Aware Shifted Windows Transformer for robust and reliable smoke detection, leveraging a novel smoke segmentation dataset, constructed via optical flow-based motion encoding, is proposed. The optical flow estimation is performed with a four-color-theorem-inspired dual-phase level-set fractional-order variational model, which preserves

arXiv:2508.14537v1 Announce Type: new Abstract: Whole slide images (WSIs) in computational pathology (CPath) pose a major computational challenge due to their gigapixel scale, often requiring the processing of tens to hundreds of thousands of high-resolution patches per slide. This results in prohibitive encoding costs, with preprocessing and training times extending to days or even weeks-making WSI encoding the most significant bottleneck in real-world deployment. In this work, we propose WISE-FUSE, an adaptive WSI encoding framework that leverages pathology-domain vision-language models and large language models to address this challenge by selectively processing diagnostically relevant regions. WISE-FUSE first computes similarity scores between low-resolution patches and class-specific textual descriptions using a knowledge distillation mechanism that preserves fine-grained diagnostic features. Based on these similarity scores, we select a small subset of informative regions for

arXiv:2508.14525v1 Announce Type: new Abstract: We introduce EffiFusion-GAN (Efficient Fusion Generative Adversarial Network), a lightweight yet powerful model for speech enhancement. The model integrates depthwise separable convolutions within a multi-scale block to capture diverse acoustic features efficiently. An enhanced attention mechanism with dual normalization and residual refinement further improves training stability and convergence. Additionally, dynamic pruning is applied to reduce model size while maintaining performance, making the framework suitable for resource-constrained environments. Experimental evaluation on the public VoiceBank+DEMAND dataset shows that EffiFusion-GAN achieves a PESQ score of 3.45, outperforming existing models under the same parameter settings.

arXiv:2508.14485v1 Announce Type: new Abstract: Traditional recommendation methods rely on correlating the embedding vectors of item IDs to capture implicit collaborative filtering signals to model the user's interest in the target item. Consequently, traditional ID-based methods often encounter data sparsity problems stemming from the sparse nature of ID features. To alleviate the problem of item ID sparsity, recommendation models incorporate multimodal item information to enhance recommendation accuracy. However, existing multimodal recommendation methods typically employ early fusion approaches, which focus primarily on combining text and image features, while neglecting the contextual influence of user behavior sequences. This oversight prevents dynamic adaptation of multimodal interest representations based on behavioral patterns, consequently restricting the model's capacity to effectively capture user multimodal interests. Therefore, this paper proposes the Distribution-Guided

arXiv:2508.14454v1 Announce Type: new Abstract: This work presents analytical solutions for the current distribution in lithium-ion battery packs composed of cells connected in parallel, explicitly accounting for the presence of interconnection resistances. These solutions enable the reformulation of the differential-algebraic equations describing the pack dynamics into a set of ordinary differential equations, thereby simplifying simulation and analysis. Conditions under which uniform current sharing across all cells occurs are also derived. The proposed formulation is validated against experimental data and confirms its ability to capture the key behaviours induced by interconnection resistances. These results can support the improved design and control of parallel-connected battery packs.

arXiv:2508.14224v1 Announce Type: new Abstract: Battery electric vehicles (BEVs) have advanced significantly during the past decade, yet drivetrain energy losses continue to restrict practical range and elevate cost. A dataset comprising more than 1000 European-market BEVs (model years 2010-2025) is combined with detailed inverter-motor co-simulation to chart technology progress for and quantify the efficiency and cost-saving potential of partial-load optimised multi-level inverter (MLI) for 2030. Average drive-cycle range has climbed from 135 km to 455 km, while fleet-average energy consumption has remained virtually constant. Three inverter topologies are assessed to evaluate future efficiency and cost enhancements: a conventional two-level (2L) six halfbridge (B6) inverter with silicon (Si) and silicon carbide (SiC) devices, and two three-level (3L) T-type neutral point clamped (TNPC) and active neutral point clamped (ANPC) inverters tailored for partial-load operation. The 3L-TNPC

Scientists from the Southwest Research Institute have found strong evidence that near-Earth asteroids Bennu and Ryugu share a common origin with Polana, a much larger asteroid in the main belt. By comparing James Webb Telescope observations with samples from NASA’s OSIRIS-REx and Japan’s Hayabusa2 missions, researchers discovered spectral similarities suggesting all three were once fragments of the same parent body, shattered in an ancient collision.

National Renewable Network CEO Alan Hunter on his company's mission to get home solar and storage to the masses. Plus news of the week. The post Solar Insiders Podcast: The “zero-cost” VPP with big plans for empty rooftops appeared first on RenewEconomy.

2GB says need for diesel generators is proof that wind energy can't keep the lights on. But network operator says it is about vulnerable poles and wires, not generation technology. The post Town to get diesel generators as back up for dodgy power lines: So 2GB lays into wind energy appeared first on RenewEconomy.

A group of homeowners worked together to navigate the process of installing rooftop solar systems, saving time and money in the process.

There's a giant solar tornado raging on the sun's surface, and a researcher captured it — plus a massive plasma eruption — in one spectacular image.

A 1989 experiment offered the promise of nuclear fusion without the need for high temperatures, but this "cold fusion" was quickly debunked. Now, some of the techniques involved have been resurrected in a new experiment that could actually improve efforts to achieve practical fusion power

The president's comment comes after the administration tightened federal permitting for renewables last month.

Ford's next cash cow might not be cars, but the batteries that underpin them.

Scientists at the University of California San Diego have uncovered how diamond—the material used to encase fuel for fusion experiments at the National Ignition Facility (NIF) in Lawrence Livermore National Laboratory—can develop tiny structural flaws that may limit fusion performance.

Civ Robotics is a startup that uses robots as land surveyors. The robots can mark thousands of placement points per day for solar panels on solar farms.

Civ Robotics is a startup that uses robots as land surveyors. The robots can mark thousands of placement points per day for solar panels on solar farms.

Using a small bench-top reactor, researchers at the University of British Columbia (UBC) have demonstrated that electrochemically loading a solid metal target with deuterium fuel can boost nuclear fusion rates.

An AI model trained on years of data from NASA’s Solar Dynamics Observatory can predict the sun’s future appearance and potentially flag dangerous solar flares

Despite losses and delays, Ford’s electric vehicle push is just getting started.

Practical fusion power that can provide cheap, clean energy could be a step closer thanks to artificial intelligence. Scientists at Lawrence Livermore National Laboratory have developed a deep learning model that accurately predicted the results of a nuclear fusion experiment conducted in 2022. Accurate predictions can help speed up the design of new experiments and accelerate the quest for this virtually limitless energy source.

AEMO's 10-year energy reliability forecast says there is enough new solar, wind and storage in the pipeline to fill the gap of retiring coal, and meet rising demand. The post AEMO says wind, solar and storage keeping energy reliability “healthy,” despite leap in data-driven demand appeared first on RenewEconomy.

NASA and IBM have released a new open-source machine learning model to help scientists better understand and predict the physics and weather patterns of the sun. Surya, trained on over a decade’s worth of NASA solar data, should help give scientists an early warning when a dangerous solar flare is likely to hit Earth. Solar…

Sometimes inspiration can strike from the most unexpected places. It can result in a cross-pollination between ideas commonly used in one field but applied to a completely different one. That might have been the case with a recent paper on lightsail design from researchers at the University of Nottingham that used techniques typically used in video games to develop a new and improved structure of a lightsail.

Windpower Monthly rounds up the latest patents for wind power technology granted and applied for in the last week.

Whether in solar cells or in the human eye, whenever certain molecules absorb light, the electrons within them shift from their ground state into a higher-energy, excited state. This results in the transport of energy and charge, leading to charge separation and eventually to the generation of electricity.

arXiv:2508.13995v1 Announce Type: new Abstract: Outside of urban hubs, autonomous cars and trucks have to master driving on intercity highways. Safe, long-distance highway travel at speeds exceeding 100 km/h demands perception distances of at least 250 m, which is about five times the 50-100m typically addressed in city driving, to allow sufficient planning and braking margins. Increasing the perception ranges also allows to extend autonomy from light two-ton passenger vehicles to large-scale forty-ton trucks, which need a longer planning horizon due to their high inertia. However, most existing perception approaches focus on shorter ranges and rely on Bird's Eye View (BEV) representations, which incur quadratic increases in memory and compute costs as distance grows. To overcome this limitation, we built on top of a sparse representation and introduced an efficient 3D encoding of multi-modal and temporal features, along with a novel self-supervised pre-training scheme that enables

arXiv:2508.13843v1 Announce Type: new Abstract: Current e-commerce multimodal retrieval systems face two key limitations: they optimize for specific tasks with fixed modality pairings, and lack comprehensive benchmarks for evaluating unified retrieval approaches. To address these challenges, we introduce UniECS, a unified multimodal e-commerce search framework that handles all retrieval scenarios across image, text, and their combinations. Our work makes three key contributions. First, we propose a flexible architecture with a novel gated multimodal encoder that uses adaptive fusion mechanisms. This encoder integrates different modality representations while handling missing modalities. Second, we develop a comprehensive training strategy to optimize learning. It combines cross-modal alignment loss (CMAL), cohesive local alignment loss (CLAL), intra-modal contrastive loss (IMCL), and adaptive loss weighting. Third, we create M-BEER, a carefully curated multimodal benchmark containing

arXiv:2508.13196v1 Announce Type: new Abstract: This paper introduces a novel approach for multimodal sentiment analysis on social media, particularly in the context of natural disasters, where understanding public sentiment is crucial for effective crisis management. Unlike conventional methods that process text and image modalities separately, our approach seamlessly integrates Convolutional Neural Network (CNN) based image analysis with Large Language Model (LLM) based text processing, leveraging Generative Pre-trained Transformer (GPT) and prompt engineering to extract sentiment relevant features from the CrisisMMD dataset. To effectively model intermodal relationships, we introduce a contextual attention mechanism within the fusion process. Leveraging contextual-attention layers, this mechanism effectively captures intermodality interactions, enhancing the model's comprehension of complex relationships between textual and visual data. The deep neural network architecture of our

Australia's first large scale solar battery hybrid is dodging weak wholesale prices and sending power into the grid in time for dinner, and often past midnight. The post The solar farm that winds down at dusk, charges up for dinner and is still generating at midnight appeared first on RenewEconomy.

Lithium battery recycling offers a powerful solution to rising demand, with discarded batteries still holding most of their valuable materials. Compared to mining, recycling slashes emissions and resource use while unlocking major economic potential. Yet infrastructure, policy, and technology hurdles must still be overcome.

For many years, Northvolt was seen as a critical part of European and American efforts to create an EV battery supply chain relatively independent of East Asia.

Good Energy Solutions is excited to announce a new partnership with Brightwell Capital, a collaboration designed to accelerate the adoption of clean energy for nonprofits, churches and government organizations. This initiative opens the door for mission-driven groups to reduce operating costs, improve energy resilience and shrink their carbon footprint, all while giving investors the chance… The post Good Energy Solutions announces new financing partnership to spur nonprofit solar installations appeared first on Solar Power World.

Accurate predictions could accelerate the design of new experiments and bring practical fusion power closer. The post This AI Model Predicts Whether Fusion Power Experiments Will Work appeared first on SingularityHub.

A major solar farm owner and operator is partnering with an electric cooperative to build a 100-MW solar power facility that will serve Meta’s first data center in South Carolina. Meta, the parent of Facebook, Instagram, and other social media and communication platforms, is expanding its relationship with Silicon Ranch, a project developer and independent […] The post Solar Farm Will Power Meta Data Center in South Carolina appeared first on POWER Magazine.

Silicon Ranch will partner with Columbia-based Central Electric Power Cooperative to build a 100-MW solar farm that will support Meta’s first data center in South Carolina. Meta’s development will support Silicon Ranch’s investment in the Orangeburg County solar farm. The 100-MW solar facility will be the fourth between Silicon Ranch and Central, a generation and… The post Meta data center in South Carolina to run on power from 100-MW Silicon Ranch solar project appeared first on Solar Power World.

German wind power developer Volkswind has selected the former head of Danish renewables developer Eurowind Energy as its new CEO after its current leader announced plans to leave the company.

arXiv:2508.12562v1 Announce Type: cross Abstract: Accurate and timely identification of pulmonary nodules on chest X-rays can differentiate between life-saving early treatment and avoidable invasive procedures. Calcification is a definitive indicator of benign nodules and is the primary foundation for diagnosis. In actual practice, diagnosing pulmonary nodule calcification on chest X-rays predominantly depends on the physician's visual assessment, resulting in significant diversity in interpretation. Furthermore, overlapping anatomical elements, such as ribs and spine, complicate the precise identification of calcification patterns. This study presents a calcification classification model that attains strong diagnostic performance by utilizing fused features derived from raw images and their structure-suppressed variants to reduce structural interference. We used 2,517 lesion-free images and 656 nodule images (151 calcified nodules and 550 non-calcified nodules), all obtained from

arXiv:2508.12048v1 Announce Type: cross Abstract: Data fusion and transfer learning are rapidly growing fields that enhance model performance for a target population by leveraging other related data sources or tasks. The challenges lie in the various potential heterogeneities between the target and external data, as well as various practical concerns that prevent a na\"ive data integration. We consider a realistic scenario where the target data is limited in size while the external data is large but contaminated with outliers; such data contamination, along with other computational and operational constraints, necessitates proper selection or subsampling of the external data for transfer learning. To our knowledge,transfer learning and subsampling under data contamination have not been thoroughly investigated. We address this gap by studying various transfer learning methods with subsamples of the external data, accounting for outliers deviating from the underlying true model due to

arXiv:2508.11666v1 Announce Type: cross Abstract: The limitations of unimodal deep learning models, particularly their tendency to overfit and limited generalizability, have renewed interest in multimodal fusion strategies. Multimodal deep neural networks (MDNN) have the capability of integrating diverse data domains and offer a promising solution for robust and accurate predictions. However, the optimal fusion strategy, intermediate fusion (feature-level) versus late fusion (decision-level) remains insufficiently examined, especially in high-stakes clinical contexts such as ECG-based cardiovascular disease (CVD) classification. This study investigates the comparative effectiveness of intermediate and late fusion strategies using ECG signals across three domains: time, frequency, and time-frequency. A series of experiments were conducted to identify the highest-performing fusion architecture. Results demonstrate that intermediate fusion consistently outperformed late fusion, achieving

arXiv:2508.13153v1 Announce Type: new Abstract: Reconstructing complete and interactive 3D scenes remains a fundamental challenge in computer vision and robotics, particularly due to persistent object occlusions and limited sensor coverage. Multiview observations from a single scene scan often fail to capture the full structural details. Existing approaches typically rely on multi stage pipelines, such as segmentation, background completion, and inpainting or require per-object dense scanning, both of which are error-prone, and not easily scalable. We propose IGFuse, a novel framework that reconstructs interactive Gaussian scene by fusing observations from multiple scans, where natural object rearrangement between captures reveal previously occluded regions. Our method constructs segmentation aware Gaussian fields and enforces bi-directional photometric and semantic consistency across scans. To handle spatial misalignments, we introduce a pseudo-intermediate scene state for unified

arXiv:2508.12917v1 Announce Type: new Abstract: Multi-modal methods based on camera and LiDAR sensors have garnered significant attention in the field of 3D detection. However, many prevalent works focus on single or partial stage fusion, leading to insufficient feature extraction and suboptimal performance. In this paper, we introduce a multi-stage cross-modal fusion 3D detection framework, termed CMF-IOU, to effectively address the challenge of aligning 3D spatial and 2D semantic information. Specifically, we first project the pixel information into 3D space via a depth completion network to get the pseudo points, which unifies the representation of the LiDAR and camera information. Then, a bilateral cross-view enhancement 3D backbone is designed to encode LiDAR points and pseudo points. The first sparse-to-distant (S2D) branch utilizes an encoder-decoder structure to reinforce the representation of sparse LiDAR points. The second residual view consistency (ResVC) branch is proposed

arXiv:2508.12750v1 Announce Type: new Abstract: Shadow removal aims to restore images that are partially degraded by shadows, where the degradation is spatially localized and non-uniform. Unlike general restoration tasks that assume global degradation, shadow removal can leverage abundant information from non-shadow regions for guidance. However, the transformation required to correct shadowed areas often differs significantly from that of well-lit regions, making it challenging to apply uniform correction strategies. This necessitates the effective integration of non-local contextual cues and adaptive modeling of region-specific transformations. To this end, we propose a novel Mamba-based network featuring dual-scale fusion and dual-path scanning to selectively propagate contextual information based on transformation similarity across regions. Specifically, the proposed Dual-Scale Fusion Mamba Block (DFMB) enhances multi-scale feature representation by fusing original features with

arXiv:2508.12526v1 Announce Type: new Abstract: The ongoing energy transition is significantly increasing the share of renewable energy sources (RES) in power systems; however, their intermittency and variability pose substantial challenges, including load shedding and system congestion. This study examines the role of the battery storage system (BSS) in mitigating these challenges by balancing power supply and demand. We optimize the location, size, and type of batteries using a two-stage stochastic program, with the second stage involving hourly operational decisions over an entire year. Unlike previous research, we incorporate the comprehensive technical and economic characteristics of battery technologies. The New York State (NYS) power system, currently undergoing a significant shift towards increased RES generation, serves as our case study. Using available load and weather data from 1980-2019, we account for the uncertainty of both load and RES generation through a sample

arXiv:2508.12484v1 Announce Type: new Abstract: Skin cancer classification is a crucial task in medical image analysis, where precise differentiation between malignant and non-malignant lesions is essential for early diagnosis and treatment. In this study, we explore Sequential and Parallel Hybrid CNN-Transformer models with Convolutional Kolmogorov-Arnold Network (CKAN). Our approach integrates transfer learning and extensive data augmentation, where CNNs extract local spatial features, Transformers model global dependencies, and CKAN facilitates nonlinear feature fusion for improved representation learning. To assess generalization, we evaluate our models on multiple benchmark datasets (HAM10000,BCN20000 and PAD-UFES) under varying data distributions and class imbalances. Experimental results demonstrate that hybrid CNN-Transformer architectures effectively capture both spatial and contextual features, leading to improved classification performance. Additionally, the integration of

arXiv:2508.12443v1 Announce Type: new Abstract: This paper presents the design, instrumentation, and experimental procedures used to test the Spherical Sailing Omnidirectional Rover (SSailOR) in a controlled wind tunnel environment. The SSailOR is a wind-powered autonomous rover. This concept is motivated by the growing need for persistent and sustainable robotic systems in applications such as planetary exploration, Arctic observation, and military surveillance. SSailOR uses wind propulsion via onboard sails to enable long-duration mobility with minimal energy consumption. The spherical design simplifies mechanical complexity while enabling omnidirectional movement. Experimental tests were conducted to validate dynamic models and assess the aerodynamic performance of the rover under various configurations and environmental conditions. As a result, this design requires a co-design approach. Details of the mechanical structure, sensor integration, electronics, data acquisition system,

arXiv:2508.12351v1 Announce Type: new Abstract: The integration of large-scale renewable energy sources, such as wind power, poses significant challenges for the optimal operation of power systems owing to their inherent uncertainties. This paper proposes a solution framework for wind-integrated optimal power flow (OPF) that leverages an enhanced second-order cone relaxation (SOCR), supported by a rolling cutting plane technique. Initially, the wind generation cost, arising from discrepancies between scheduled and actual wind power outputs, is meticulously modeled using a Gaussian mixture model based on historical wind power data. This modelled wind generation cost is subsequently incorporated into the objective function of the conventional OPF problem. To achieve the efficient and accurate solution for the wind-integrated OPF, effectively managing the constraints associated with AC power flow equations is essential. In this regard, a SOCR, combined with a second-order Taylor series

arXiv:2508.12036v1 Announce Type: new Abstract: Solving tough clinical questions that require both image and text understanding is still a major challenge in healthcare AI. In this work, we propose Q-FSRU, a new model that combines Frequency Spectrum Representation and Fusion (FSRU) with a method called Quantum Retrieval-Augmented Generation (Quantum RAG) for medical Visual Question Answering (VQA). The model takes in features from medical images and related text, then shifts them into the frequency domain using Fast Fourier Transform (FFT). This helps it focus on more meaningful data and filter out noise or less useful information. To improve accuracy and ensure that answers are based on real knowledge, we add a quantum-inspired retrieval system. It fetches useful medical facts from external sources using quantum-based similarity techniques. These details are then merged with the frequency-based features for stronger reasoning. We evaluated our model using the VQA-RAD dataset, which

arXiv:2508.11732v1 Announce Type: new Abstract: Existing deep learning models for functional MRI-based classification have limitations in network architecture determination (relying on experience) and feature space fusion (mostly simple concatenation, lacking mutual learning). Inspired by the human brain's mechanism of updating neural connections through learning and decision-making, we proposed a novel BRain-Inspired feature Fusion (BRIEF) framework, which is able to optimize network architecture automatically by incorporating an improved neural network connection search (NCS) strategy and a Transformer-based multi-feature fusion module. Specifically, we first extracted 4 types of fMRI temporal representations, i.e., time series (TCs), static/dynamic functional connection (FNC/dFNC), and multi-scale dispersion entropy (MsDE), to construct four encoders. Within each encoder, we employed a modified Q-learning to dynamically optimize the NCS to extract high-level feature vectors, where

HMC seeks $1 billion in new attempt at fund raising its Gillard-chaired energy platform, but it has narrowed its development focus to wind and battery storage. The post HMC narrows focus to wind and big batteries as it tries again to raise $1 billion for new energy platform appeared first on RenewEconomy.

Got a particle accelerator? Here’s your tritium startup idea Tritium is ridiculously rare, incredibly expensive, and central to most fusion energy reactor designs. If research out of Los Alamos National Lab proves to hold true, it might soon become easier to obtain.…

A research team affiliated with UNIST has unveiled a technology that transforms nitrates found in wastewater into ammonia, a vital chemical and promising energy carrier, without carbon emissions. This advancement not only offers a sustainable method for ammonia production but also contributes to wastewater purification efforts.

New research led by Southwest Research Institute (SwRI) has confirmed decades-old theoretical models of magnetic reconnection, the process that releases stored magnetic energy to drive solar flares, coronal mass ejections and other space weather phenomena. The data was captured by NASA's Parker Solar Probe (PSP), which is the only spacecraft to have flown through the sun's upper atmosphere.

P2P power networks beat stingy feed-in tariffs for Aussie households, study finds Boffins looking into the Australian solar energy ecosystem say that sharing really is caring – and potentially profitable when homes with solar panels can sell their excess energy to neighbors at a preferential rate.…

Researchers from Macquarie University claim a breakthrough in efforts to extract valuable silver from discarded solar modules without destroying other panel components. The post Australia researchers find silver lining in breakthrough technology for discarded solar panels appeared first on RenewEconomy.

Japanese offshore wind developer ENEOS Renewable Energy (ERE) has entered into an agreement with […]

The US treasury has tightened the conditions for wind and solar power projects to receive renewable energy tax credits before they expire in a further blow to the country’s wind industry. 

The gray box mounted next to your garage does more than convert sunlight into electricity. That solar inverter

arXiv:2508.11375v1 Announce Type: cross Abstract: Medical semantic-mask synthesis boosts data augmentation and analysis, yet most GAN-based approaches still produce one-to-one images and lack spatial consistency in complex scans. To address this, we propose AnatoMaskGAN, a novel synthesis framework that embeds slice-related spatial features to precisely aggregate inter-slice contextual dependencies, introduces diverse image-augmentation strategies, and optimizes deep feature learning to improve performance on complex medical images. Specifically, we design a GNN-based strongly correlated slice-feature fusion module to model spatial relationships between slices and integrate contextual information from neighboring slices, thereby capturing anatomical details more comprehensively; we introduce a three-dimensional spatial noise-injection strategy that weights and fuses spatial features with noise to enhance modeling of structural diversity; and we incorporate a grayscale-texture

arXiv:2508.11259v1 Announce Type: cross Abstract: This paper proposes a novel spatiotemporal (ST) fusion framework for satellite images, named Temporally-Similar Structure-Aware ST fusion (TSSTF). ST fusion is a promising approach to address the trade-off between the spatial and temporal resolution of satellite images. In real-world scenarios, observed satellite images are severely degraded by noise due to measurement equipment and environmental conditions. Consequently, some recent studies have focused on enhancing the robustness of ST fusion methods against noise. However, existing noise-robust ST fusion approaches often fail to capture fine spatial structure, leading to oversmoothing and artifacts. To address this issue, TSSTF introduces two key mechanisms: Temporally-Guided Total Variation (TGTV) and Temporally-Guided Edge Constraint (TGEC). TGTV is a novel regularization function that promotes spatial piecewise smoothness while preserving structural details, guided by a reference

arXiv:2508.11485v1 Announce Type: new Abstract: Accurate and reliable navigation is crucial for autonomous unmanned ground vehicle (UGV). However, current UGV datasets fall short in meeting the demands for advancing navigation and mapping techniques due to limitations in sensor configuration, time synchronization, ground truth, and scenario diversity. To address these challenges, we present i2Nav-Robot, a large-scale dataset designed for multi-sensor fusion navigation and mapping in indoor-outdoor environments. We integrate multi-modal sensors, including the newest front-view and 360-degree solid-state LiDARs, 4-dimensional (4D) radar, stereo cameras, odometer, global navigation satellite system (GNSS) receiver, and inertial measurement units (IMU) on an omnidirectional wheeled robot. Accurate timestamps are obtained through both online hardware synchronization and offline calibration for all sensors. The dataset comprises ten larger-scale sequences covering diverse UGV operating