Efficient and accurate whole-brain lesion segmentation remains a challenge in medical image analysis. In this work, we revisit MeshNet, a parameter-efficient segmentation model, and introduce a novel multi-scale dilation pattern with an encoder-decoder structure. This innovation enables capturing broad contextual information and fine-grained details without traditional downsampling, upsampling, or skip-connections. Unlike previous approaches processing subvolumes or slices, we operate directly on whole-brain $256^3$ MRI volumes. Evaluations on the Aphasia Recovery Cohort (ARC) dataset demonstrate that MeshNet achieves superior or comparable DICE scores to state-of-the-art architectures such as MedNeXt and U-MAMBA at 1/1000th of parameters. Our results validate MeshNet's strong balance of efficiency and performance, making it particularly suitable for resource-limited environments such as web-based applications and opening new possibilities for the widespread deployment of advanced medical image analysis tools.
We investigate methods to reduce inference time and memory footprint in stable diffusion models by introducing lightweight decoders for both image and video synthesis. Traditional latent diffusion pipelines rely on large Variational Autoencoder decoders that can slow down generation and consume considerable GPU memory. We propose custom-trained decoders using lightweight Vision Transformer and Taming Transformer architectures. Experiments show up to 15% overall speed-ups for image generation on COCO2017 and up to 20 times faster decoding in the sub-module, with additional gains on UCF-101 for video tasks. Memory requirements are moderately reduced, and while there is a small drop in perceptual quality compared to the default decoder, the improvements in speed and scalability are crucial for large-scale inference scenarios such as generating 100K images. Our work is further contextualized by advances in efficient video generation, including dual masking strategies, illustrating a broader effort to improve the scalability and efficiency of generative models.
Designing generative models for 3D structural brain MRI that synthesize morphologically-plausible and attribute-specific (e.g., age, sex, disease state) samples is an active area of research. Existing approaches based on frameworks like GANs or diffusion models synthesize the image directly, which may limit their ability to capture intricate morphological details. In this work, we propose a 3D brain MRI generation method based on state-of-the-art latent diffusion models (LDMs), called MorphLDM, that generates novel images by applying synthesized deformation fields to a learned template. Instead of using a reconstruction-based autoencoder (as in a typical LDM), our encoder outputs a latent embedding derived from both an image and a learned template that is itself the output of a template decoder; this latent is passed to a deformation field decoder, whose output is applied to the learned template. A registration loss is minimized between the original image and the deformed template with respect to the encoder and both decoders. Empirically, our approach outperforms generative baselines on metrics spanning image diversity, adherence with respect to input conditions, and voxel-based morphometry. Our code is available at https://github.com/alanqrwang/morphldm.
Using the information theory, this study provides insights into how the construction of latent space of autoencoder (AE) using deep neural network (DNN) training finds a smooth low-dimensional manifold in the stiff dynamical system. Our recent study [1] reported that an autoencoder (AE) combined with neural ODE (NODE) as a surrogate reduced order model (ROM) for the integration of stiff chemically reacting systems led to a significant reduction in the temporal stiffness, and the behavior was attributed to the identification of a slow invariant manifold by the nonlinear projection of the AE. The present work offers fundamental understanding of the mechanism by employing concepts from information theory and better mixing. The learning mechanism of both the encoder and decoder are explained by plotting the evolution of mutual information and identifying two different phases. Subsequently, the density distribution is plotted for the physical and latent variables, which shows the transformation of the \emph{rare event} in the physical space to a \emph{highly likely} (more probable) event in the latent space provided by the nonlinear autoencoder. Finally, the nonlinear transformation leading to density redistribution is explained using concepts from information theory and probability.
Quantum computing leverages qubits, exploiting superposition and entanglement to solve problems intractable for classical computers, offering significant computational advantages. Quantum machine learning (QML), which integrates quantum computing with machine learning, holds immense potential across various fields but remains largely unexplored in geosciences. However, its progress is hindered by the limitations of current NISQ hardware. To address these challenges, hybrid quantum neural networks (HQNNs) have emerged, combining quantum layers within classical neural networks to leverage the strengths of both paradigms. To the best of our knowledge, this study presents the first application of QML to subsurface imaging through the development of hybrid quantum physics-informed neural networks (HQ-PINNs) for seismic inversion. We apply the HQ-PINN framework to invert pre-stack and post-stack seismic datasets, estimating P- and S-impedances. The proposed HQ-PINN architecture follows an encoder-decoder structure, where the encoder (HQNN), processes seismic data to estimate elastic parameters, while the decoder utilizes these parameters to generate the corresponding seismic data based on geophysical relationships. The HQ-PINN model is trained by minimizing the misfit between the input and predicted seismic data generated by the decoder. We systematically evaluate various quantum layer configurations, differentiation methods, and quantum device simulators on the inversion performance, and demonstrate real-world applicability through the individual and simultaneous inversion cases of the Sleipner dataset. The HQ-PINN framework consistently and efficiently estimated accurate subsurface impedances across the synthetic and field case studies, establishing the feasibility of leveraging QML for seismic inversion, thereby paving the way for broader applications of quantum computing in geosciences.
Diverse usage patterns induce complex and variable aging behaviors in lithium-ion batteries, complicating accurate health diagnosis and prognosis. Separate diagnostic cycles are often used to untangle the battery's current state of health from prior complex aging patterns. However, these same diagnostic cycles alter the battery's degradation trajectory, are time-intensive, and cannot be practically performed in onboard applications. In this work, we leverage portions of operational measurements in combination with an interpretable machine learning model to enable rapid, onboard battery health diagnostics and prognostics without offline diagnostic testing and the requirement of historical data. We integrate mechanistic constraints within an encoder-decoder architecture to extract electrode states in a physically interpretable latent space and enable improved reconstruction of the degradation path. The health diagnosis model framework can be flexibly applied across diverse application interests with slight fine-tuning. We demonstrate the versatility of this model framework by applying it to three battery-cycling datasets consisting of 422 cells under different operating conditions, highlighting the utility of an interpretable diagnostic-free, onboard battery diagnosis and prognosis model.
This paper introduces PoseLess, a novel framework for robot hand control that eliminates the need for explicit pose estimation by directly mapping 2D images to joint angles using projected representations. Our approach leverages synthetic training data generated through randomized joint configurations, enabling zero-shot generalization to real-world scenarios and cross-morphology transfer from robotic to human hands. By projecting visual inputs and employing a transformer-based decoder, PoseLess achieves robust, low-latency control while addressing challenges such as depth ambiguity and data scarcity. Experimental results demonstrate competitive performance in joint angle prediction accuracy without relying on any human-labelled dataset.
Multi-modal emotion recognition is challenging due to the difficulty of extracting features that capture subtle emotional differences. Understanding multi-modal interactions and connections is key to building effective bimodal speech emotion recognition systems. In this work, we propose Bimodal Connection Attention Fusion (BCAF) method, which includes three main modules: the interactive connection network, the bimodal attention network, and the correlative attention network. The interactive connection network uses an encoder-decoder architecture to model modality connections between audio and text while leveraging modality-specific features. The bimodal attention network enhances semantic complementation and exploits intra- and inter-modal interactions. The correlative attention network reduces cross-modal noise and captures correlations between audio and text. Experiments on the MELD and IEMOCAP datasets demonstrate that the proposed BCAF method outperforms existing state-of-the-art baselines.
Decoder-based transformers, while revolutionizing language modeling and scaling to immense sizes, have not completely overtaken encoder-heavy architectures in natural language processing. Specifically, encoder-only models remain dominant in tasks like classification, regression, and ranking. This is primarily due to the inherent structure of decoder-based models, which limits their direct applicability to these tasks. In this paper, we introduce Gemma Encoder, adapting the powerful Gemma decoder model to an encoder architecture, thereby unlocking its potential for a wider range of non-generative applications. To optimize the adaptation from decoder to encoder, we systematically analyze various pooling strategies, attention mechanisms, and hyperparameters (e.g., dropout rate). Furthermore, we benchmark Gemma Encoder against established approaches on the GLUE benchmarks, and MS MARCO ranking benchmark, demonstrating its effectiveness and versatility.
Existing video tokenizers typically use the traditional Variational Autoencoder (VAE) architecture for video compression and reconstruction. However, to achieve good performance, its training process often relies on complex multi-stage training tricks that go beyond basic reconstruction loss and KL regularization. Among these tricks, the most challenging is the precise tuning of adversarial training with additional Generative Adversarial Networks (GANs) in the final stage, which can hinder stable convergence. In contrast to GANs, diffusion models offer more stable training processes and can generate higher-quality results. Inspired by these advantages, we propose CDT, a novel Conditioned Diffusion-based video Tokenizer, that replaces the GAN-based decoder with a conditional causal diffusion model. The encoder compresses spatio-temporal information into compact latents, while the decoder reconstructs videos through a reverse diffusion process conditioned on these latents. During inference, we incorporate a feature cache mechanism to generate videos of arbitrary length while maintaining temporal continuity and adopt sampling acceleration technique to enhance efficiency. Trained using only a basic MSE diffusion loss for reconstruction, along with KL term and LPIPS perceptual loss from scratch, extensive experiments demonstrate that CDT achieves state-of-the-art performance in video reconstruction tasks with just a single-step sampling. Even a scaled-down version of CDT (3$\times$ inference speedup) still performs comparably with top baselines. Moreover, the latent video generation model trained with CDT also exhibits superior performance. The source code and pretrained weights are available at https://github.com/ali-vilab/CDT.
Bangla, a language spoken by over 300 million native speakers and ranked as the sixth most spoken language worldwide, presents unique challenges in natural language processing (NLP) due to its complex morphological characteristics and limited resources. While recent Large Decoder Based models (LLMs), such as GPT, LLaMA, and DeepSeek, have demonstrated excellent performance across many NLP tasks, their effectiveness in Bangla remains largely unexplored. In this paper, we establish the first benchmark comparing decoder-based LLMs with classic encoder-based models for Zero-Shot Multi-Label Classification (Zero-Shot-MLC) task in Bangla. Our evaluation of 32 state-of-the-art models reveals that, existing so-called powerful encoders and decoders still struggle to achieve high accuracy on the Bangla Zero-Shot-MLC task, suggesting a need for more research and resources for Bangla NLP.
Accurate time series prediction is challenging due to the inherent nonlinearity and sensitivity to initial conditions. We propose a novel approach that enhances neural network predictions through differential learning, which involves training models on both the original time series and its differential series. Specifically, we develop a differential long short-term memory (Diff-LSTM) network that uses a shared LSTM cell to simultaneously process both data streams, effectively capturing intrinsic patterns and temporal dynamics. Evaluated on the Mackey-Glass, Lorenz, and R\"ossler chaotic time series, as well as a real-world financial dataset from ACI Worldwide Inc., our results demonstrate that the Diff- LSTM network outperforms prevalent models such as recurrent neural networks, convolutional neural networks, and bidirectional and encoder-decoder LSTM networks in both short-term and long-term predictions. This framework offers a promising solution for enhancing time series prediction, even when comprehensive knowledge of the underlying dynamics of the time series is not fully available.
Melanoma is a malignant tumor originating from skin cell lesions. Accurate and efficient segmentation of skin lesions is essential for quantitative medical analysis but remains challenging. To address this, we propose ScaleFusionNet, a segmentation model that integrates Cross-Attention Transformer Module (CATM) and AdaptiveFusionBlock to enhance feature extraction and fusion. The model employs a hybrid architecture encoder that effectively captures both local and global features. We introduce CATM, which utilizes Swin Transformer Blocks and Cross Attention Fusion (CAF) to adaptively refine encoder-decoder feature fusion, reducing semantic gaps and improving segmentation accuracy. Additionally, the AdaptiveFusionBlock is improved by integrating adaptive multi-scale fusion, where Swin Transformer-based attention complements deformable convolution-based multi-scale feature extraction. This enhancement refines lesion boundaries and preserves fine-grained details. ScaleFusionNet achieves Dice scores of 92.94% and 91.65% on ISIC-2016 and ISIC-2018 datasets, respectively, demonstrating its effectiveness in skin lesion analysis. Our code implementation is publicly available at GitHub.
Image segmentation is a fundamental task in both image analysis and medical applications. State-of-the-art methods predominantly rely on encoder-decoder architectures with a U-shaped design, commonly referred to as U-Net. Recent advancements integrating transformers and MLPs improve performance but still face key limitations, such as poor interpretability, difficulty handling intrinsic noise, and constrained expressiveness due to discrete layer structures, often lacking a solid theoretical foundation.In this work, we introduce Implicit U-KAN 2.0, a novel U-Net variant that adopts a two-phase encoder-decoder structure. In the SONO phase, we use a second-order neural ordinary differential equation (NODEs), called the SONO block, for a more efficient, expressive, and theoretically grounded modeling approach. In the SONO-MultiKAN phase, we integrate the second-order NODEs and MultiKAN layer as the core computational block to enhance interpretability and representation power. Our contributions are threefold. First, U-KAN 2.0 is an implicit deep neural network incorporating MultiKAN and second order NODEs, improving interpretability and performance while reducing computational costs. Second, we provide a theoretical analysis demonstrating that the approximation ability of the MultiKAN block is independent of the input dimension. Third, we conduct extensive experiments on a variety of 2D and a single 3D dataset, demonstrating that our model consistently outperforms existing segmentation networks.
The perception capability of robotic systems relies on the richness of the dataset. Although Segment Anything Model 2 (SAM2), trained on large datasets, demonstrates strong perception potential in perception tasks, its inherent training paradigm prevents it from being suitable for RGB-T tasks. To address these challenges, we propose SHIFNet, a novel SAM2-driven Hybrid Interaction Paradigm that unlocks the potential of SAM2 with linguistic guidance for efficient RGB-Thermal perception. Our framework consists of two key components: (1) Semantic-Aware Cross-modal Fusion (SACF) module that dynamically balances modality contributions through text-guided affinity learning, overcoming SAM2's inherent RGB bias; (2) Heterogeneous Prompting Decoder (HPD) that enhances global semantic information through a semantic enhancement module and then combined with category embeddings to amplify cross-modal semantic consistency. With 32.27M trainable parameters, SHIFNet achieves state-of-the-art segmentation performance on public benchmarks, reaching 89.8% on PST900 and 67.8% on FMB, respectively. The framework facilitates the adaptation of pre-trained large models to RGB-T segmentation tasks, effectively mitigating the high costs associated with data collection while endowing robotic systems with comprehensive perception capabilities. The source code will be made publicly available at https://github.com/iAsakiT3T/SHIFNet.
The Entrance Dependent Vehicle Routing Problem (EDVRP) is a variant of the Vehicle Routing Problem (VRP) where the scale of cities influences routing outcomes, necessitating consideration of their entrances. This paper addresses EDVRP in agriculture, focusing on multi-parameter vehicle planning for irregularly shaped fields. To address the limitations of traditional methods, such as heuristic approaches, which often overlook field geometry and entrance constraints, we propose a Joint Probability Distribution Sampling Neural Network (JPDS-NN) to effectively solve the EDVRP. The network uses an encoder-decoder architecture with graph transformers and attention mechanisms to model routing as a Markov Decision Process, and is trained via reinforcement learning for efficient and rapid end-to-end planning. Experimental results indicate that JPDS-NN reduces travel distances by 48.4-65.4%, lowers fuel consumption by 14.0-17.6%, and computes two orders of magnitude faster than baseline methods, while demonstrating 15-25% superior performance in dynamic arrangement scenarios. Ablation studies validate the necessity of cross-attention and pre-training. The framework enables scalable, intelligent routing for large-scale farming under dynamic constraints.
We introduce ArcPro, a novel learning framework built on architectural programs to recover structured 3D abstractions from highly sparse and low-quality point clouds. Specifically, we design a domain-specific language (DSL) to hierarchically represent building structures as a program, which can be efficiently converted into a mesh. We bridge feedforward and inverse procedural modeling by using a feedforward process for training data synthesis, allowing the network to make reverse predictions. We train an encoder-decoder on the points-program pairs to establish a mapping from unstructured point clouds to architectural programs, where a 3D convolutional encoder extracts point cloud features and a transformer decoder autoregressively predicts the programs in a tokenized form. Inference by our method is highly efficient and produces plausible and faithful 3D abstractions. Comprehensive experiments demonstrate that ArcPro outperforms both traditional architectural proxy reconstruction and learning-based abstraction methods. We further explore its potential to work with multi-view image and natural language inputs.
Scribble annotations significantly reduce the cost and labor required for dense labeling in large medical datasets with complex anatomical structures. However, current scribble-supervised learning methods are limited in their ability to effectively propagate sparse annotation labels to dense segmentation masks and accurately segment object boundaries. To address these issues, we propose a Progressive Collaborative Learning framework that leverages novel algorithms and the Med-SAM foundation model to enhance information quality during training. (1) We enrich ground truth scribble segmentation labels through a new algorithm, propagating scribbles to estimate object boundaries. (2) We enhance feature representation by optimizing Med-SAM-guided training through the fusion of feature embeddings from Med-SAM and our proposed Sparse Mamba network. This enriched representation also facilitates the fine-tuning of the Med-SAM decoder with enriched scribbles. (3) For inference, we introduce a Sparse Mamba network, which is highly capable of capturing local and global dependencies by replacing the traditional sequential patch processing method with a skip-sampling procedure. Experiments on the ACDC, CHAOS, and MSCMRSeg datasets validate the effectiveness of our framework, outperforming nine state-of-the-art methods. Our code is available at \href{https://github.com/QLYCode/SparseMamba-PCL}{SparseMamba-PCL.git}.
Image dehazing is a crucial task that involves the enhancement of degraded images to recover their sharpness and textures. While vision Transformers have exhibited impressive results in diverse dehazing tasks, their quadratic complexity and lack of dehazing priors pose significant drawbacks for real-world applications. In this paper, guided by triple priors, Bright Channel Prior (BCP), Dark Channel Prior (DCP), and Histogram Equalization (HE), we propose a \textit{P}rior-\textit{g}uided Hierarchical \textit{H}armonization Network (PGH$^2$Net) for image dehazing. PGH$^2$Net is built upon the UNet-like architecture with an efficient encoder and decoder, consisting of two module types: (1) Prior aggregation module that injects B/DCP and selects diverse contexts with gating attention. (2) Feature harmonization modules that subtract low-frequency components from spatial and channel aspects and learn more informative feature distributions to equalize the feature maps.
This paper describes a streaming audio-to-MIDI piano transcription approach that aims to sequentially translate a music signal into a sequence of note onset and offset events. The sequence-to-sequence nature of this task may call for the computationally-intensive transformer model for better performance, which has recently been used for offline transcription benchmarks and could be extended for streaming transcription with causal attention mechanisms. We assume that the performance limitation of this naive approach lies in the decoder. Although time-frequency features useful for onset detection are considerably different from those for offset detection, the single decoder is trained to output a mixed sequence of onset and offset events without guarantee of the correspondence between the onset and offset events of the same note. To overcome this limitation, we propose a streaming encoder-decoder model that uses a convolutional encoder aggregating local acoustic features, followed by an autoregressive Transformer decoder detecting a variable number of onset events and another decoder detecting the offset events for the active pitches with validation of the sustain pedal at each time frame. Experiments using the MAESTRO dataset showed that the proposed streaming method performed comparably with or even better than the state-of-the-art offline methods while significantly reducing the computational cost.