MoLe-VLA: Dynamic Layer-skipping Vision Language Action Model via Mixture-of-Layers for Efficient Robot Manipulation

Multimodal Large Language Models (MLLMs) excel in understanding complex language and visual data, enabling generalist robotic systems to interpret instructions and perform embodied tasks. Nevertheless, their real-world deployment is hindered by substantial computational and storage demands. Recent insights into the homogeneous patterns in the LLM layer have inspired sparsification techniques to address these challenges, such as early exit and token pruning. However, these methods often neglect the critical role of the final layers that encode the semantic information most relevant to downstream robotic tasks. Aligning with the recent breakthrough of the Shallow Brain Hypothesis (SBH) in neuroscience and the mixture of experts in model sparsification, we conceptualize each LLM layer as an expert and propose a Mixture-of-Layers Vision-Language-Action model (MoLe-VLA, or simply MoLe) architecture for dynamic LLM layer activation. We introduce a Spatial-Temporal Aware Router (STAR) for MoLe to selectively activate only parts of the layers based on the robot's current state, mimicking the brain's distinct signal pathways specialized for cognition and causal reasoning. Additionally, to compensate for the cognitive ability of LLMs lost in MoLe, we devise a Cognition Self-Knowledge Distillation (CogKD) framework. CogKD enhances the understanding of task demands and improves the generation of task-relevant action sequences by leveraging cognitive features. Extensive experiments conducted in both RLBench simulation and real-world environments demonstrate the superiority of MoLe-VLA in both efficiency and performance. Specifically, MoLe-VLA achieves an 8% improvement in the mean success rate across ten tasks while reducing computational costs by up to x5.6 compared to standard LLMs.

StyleMorpheus: A Style-Based 3D-Aware Morphable Face Model

For 3D face modeling, the recently developed 3D-aware neural rendering methods are able to render photorealistic face images with arbitrary viewing directions. The training of the parametric controllable 3D-aware face models, however, still relies on a large-scale dataset that is lab-collected. To address this issue, this paper introduces "StyleMorpheus", the first style-based neural 3D Morphable Face Model (3DMM) that is trained on in-the-wild images. It inherits 3DMM's disentangled controllability (over face identity, expression, and appearance) but without the need for accurately reconstructed explicit 3D shapes. StyleMorpheus employs an auto-encoder structure. The encoder aims at learning a representative disentangled parametric code space and the decoder improves the disentanglement using shape and appearance-related style codes in the different sub-modules of the network. Furthermore, we fine-tune the decoder through style-based generative adversarial learning to achieve photorealistic 3D rendering quality. The proposed style-based design enables StyleMorpheus to achieve state-of-the-art 3D-aware face reconstruction results, while also allowing disentangled control of the reconstructed face. Our model achieves real-time rendering speed, allowing its use in virtual reality applications. We also demonstrate the capability of the proposed style-based design in face editing applications such as style mixing and color editing. Project homepage: https://github.com/ubc-3d-vision-lab/StyleMorpheus.

VVRec: Reconstruction Attacks on DL-based Volumetric Video Upstreaming via Latent Diffusion Model with Gamma Distribution

With the popularity of 3D volumetric video applications, such as Autonomous Driving, Virtual Reality, and Mixed Reality, current developers have turned to deep learning for compressing volumetric video frames, i.e., point clouds for video upstreaming. The latest deep learning-based solutions offer higher efficiency, lower distortion, and better hardware support compared to traditional ones like MPEG and JPEG. However, privacy threats arise, especially reconstruction attacks targeting to recover the original input point cloud from the intermediate results. In this paper, we design VVRec, to the best of our knowledge, which is the first targeting DL-based Volumetric Video Reconstruction attack scheme. VVRec demonstrates the ability to reconstruct high-quality point clouds from intercepted transmission intermediate results using four well-trained neural network modules we design. Leveraging the latest latent diffusion models with Gamma distribution and a refinement algorithm, VVRec excels in reconstruction quality, color recovery, and surpasses existing defenses. We evaluate VVRec using three volumetric video datasets. The results demonstrate that VVRec achieves 64.70dB reconstruction accuracy, with an impressive 46.39% reduction of distortion over baselines.

Mitigating Data Scarcity in Time Series Analysis: A Foundation Model with Series-Symbol Data Generation

Foundation models for time series analysis (TSA) have attracted significant attention. However, challenges such as data scarcity and data imbalance continue to hinder their development. To address this, we consider modeling complex systems through symbolic expressions that serve as semantic descriptors of time series. Building on this concept, we introduce a series-symbol (S2) dual-modulity data generation mechanism, enabling the unrestricted creation of high-quality time series data paired with corresponding symbolic representations. Leveraging the S2 dataset, we develop SymTime, a pre-trained foundation model for TSA. SymTime demonstrates competitive performance across five major TSA tasks when fine-tuned with downstream task, rivaling foundation models pre-trained on real-world datasets. This approach underscores the potential of dual-modality data generation and pretraining mechanisms in overcoming data scarcity and enhancing task performance.

Self-adaptive vision-language model for 3D segmentation of pulmonary artery and vein

Accurate segmentation of pulmonary structures iscrucial in clinical diagnosis, disease study, and treatment planning. Significant progress has been made in deep learning-based segmentation techniques, but most require much labeled data for training. Consequently, developing precise segmentation methods that demand fewer labeled datasets is paramount in medical image analysis. The emergence of pre-trained vision-language foundation models, such as CLIP, recently opened the door for universal computer vision tasks. Exploiting the generalization ability of these pre-trained foundation models on downstream tasks, such as segmentation, leads to unexpected performance with a relatively small amount of labeled data. However, exploring these models for pulmonary artery-vein segmentation is still limited. This paper proposes a novel framework called Language-guided self-adaptive Cross-Attention Fusion Framework. Our method adopts pre-trained CLIP as a strong feature extractor for generating the segmentation of 3D CT scans, while adaptively aggregating the cross-modality of text and image representations. We propose a s pecially designed adapter module to fine-tune pre-trained CLIP with a self-adaptive learning strategy to effectively fuse the two modalities of embeddings. We extensively validate our method on a local dataset, which is the largest pulmonary artery-vein CT dataset to date and consists of 718 labeled data in total. The experiments show that our method outperformed other state-of-the-art methods by a large margin. Our data and code will be made publicly available upon acceptance.

Enabling Time-series Foundation Model for Building Energy Forecasting via Contrastive Curriculum Learning

Advances in time-series forecasting are driving a shift from conventional machine learning models to foundation models (FMs) that are trained with generalized knowledge. However, existing FMs still perform poorly in the energy fields, such as building energy forecasting (BEF). This paper studies the adaptation of FM to BEF tasks. We demonstrate the shortcomings of fine-tuning FM straightforwardly from both the perspectives of FM and the data. To overcome these limitations, we propose a new \textit{contrastive curriculum learning}-based training method. Our method optimizes the ordering of training data in the context of TSFM adaptation. Experiments show that our method can improve the zero/few-shot performance by 14.6\% compared to the existing FMs. Our code and new TSFM will be available at <Anonymous Github Repo>.

Reference Signal-Based Waveform Design for Integrated Sensing and Communications System

Integrated sensing and communications (ISAC) as one of the key technologies is capable of supporting high-speed communication and high-precision sensing for the upcoming 6G. This paper studies a waveform strategy by designing the orthogonal frequency division multiplexing (OFDM)-based reference signal (RS) for sensing and communication in ISAC system. We derive the closed-form expressions of Cram\'er-Rao Bound (CRB) for the distance and velocity estimations, and obtain the communication rate under the mean square error of channel estimation. Then, a weighted sum CRB minimization problem on the distance and velocity estimations is formulated by considering communication rate requirement and RS intervals constraints, which is a mixed-integer problem due to the discrete RS interval values. To solve this problem, some numerical methods are typically adopted to obtain the optimal solutions, whose computational complexity grow exponentially with the number of symbols and subcarriers of OFDM. Therefore, we propose a relaxation and approximation method to transform the original discrete problem into a continuous convex one and obtain the sub-optimal solutions. Finally, our proposed scheme is compared with the exhaustive search method in numerical simulations, which show slight gap between the obtained sub-optimal and optimal solutions, and this gap further decreases with large weight factor.

MMM-RS: A Multi-modal, Multi-GSD, Multi-scene Remote Sensing Dataset and Benchmark for Text-to-Image Generation

Recently, the diffusion-based generative paradigm has achieved impressive general image generation capabilities with text prompts due to its accurate distribution modeling and stable training process. However, generating diverse remote sensing (RS) images that are tremendously different from general images in terms of scale and perspective remains a formidable challenge due to the lack of a comprehensive remote sensing image generation dataset with various modalities, ground sample distances (GSD), and scenes. In this paper, we propose a Multi-modal, Multi-GSD, Multi-scene Remote Sensing (MMM-RS) dataset and benchmark for text-to-image generation in diverse remote sensing scenarios. Specifically, we first collect nine publicly available RS datasets and conduct standardization for all samples. To bridge RS images to textual semantic information, we utilize a large-scale pretrained vision-language model to automatically output text prompts and perform hand-crafted rectification, resulting in information-rich text-image pairs (including multi-modal images). In particular, we design some methods to obtain the images with different GSD and various environments (e.g., low-light, foggy) in a single sample. With extensive manual screening and refining annotations, we ultimately obtain a MMM-RS dataset that comprises approximately 2.1 million text-image pairs. Extensive experimental results verify that our proposed MMM-RS dataset allows off-the-shelf diffusion models to generate diverse RS images across various modalities, scenes, weather conditions, and GSD. The dataset is available at https://github.com/ljl5261/MMM-RS.

Alt-MoE: Multimodal Alignment via Alternating Optimization of Multi-directional MoE with Unimodal Models

Recent Large Multi-Modal Models (LMMs) have made significant advancements in multi-modal alignment by employing lightweight connection modules to facilitate the representation and fusion of knowledge from existing pre-trained uni-modal models. However, these methods still rely on modality-specific and direction-specific connectors, leading to compartmentalized knowledge representations and reduced computational efficiency, which limits the model's ability to form unified multi-modal representations. To address these issues, we introduce a novel training framework, Alt-MoE, which employs the Mixture of Experts (MoE) as a unified multi-directional connector across modalities, and employs a multi-step sequential alternating unidirectional alignment strategy, which converges to bidirectional alignment over iterations. The extensive empirical studies revealed the following key points: 1) Alt-MoE achieves competitive results by integrating diverse knowledge representations from uni-modal models. This approach seamlessly fuses the specialized expertise of existing high-performance uni-modal models, effectively synthesizing their domain-specific knowledge into a cohesive multi-modal representation. 2) Alt-MoE efficiently scales to new tasks and modalities without altering its model architecture or training strategy. Furthermore, Alt-MoE operates in latent space, supporting vector pre-storage and real-time retrieval via lightweight multi-directional MoE, thereby facilitating massive data processing. Our methodology has been validated on several well-performing uni-modal models (LLAMA3, Qwen2, and DINOv2), achieving competitive results on a wide range of downstream tasks and datasets.

Receiver Selection and Transmit Beamforming for Multi-static Integrated Sensing and Communications

Next-generation wireless networks are expected to develop a novel paradigm of integrated sensing and communications (ISAC) to enable both the high-accuracy sensing and high-speed communications. However, conventional mono-static ISAC systems, which simultaneously transmit and receive at the same equipment, may suffer from severe self-interference, and thus significantly degrade the system performance.To address this issue, this paper studies a multi-static ISAC system for cooperative target localization and communications, where the transmitter transmits ISAC signal to multiple receivers (REs) deployed at different positions. We derive the closed-form Cram\'{e}r-Rao bound (CRB) on the joint estimations of both the transmission delay and Doppler shift for cooperative target localization, and the CRB minimization problem is formulated by considering the cooperative cost and communication rate requirements for the REs. To solve this problem, we first decouple it into two subproblems for RE selection and transmit beamforming, respectively. Then, a minimax linkage-based method is proposed to solve the RE selection subproblem, and a successive convex approximation algorithm is adopted to deal with the transmit beamforming subproblem with non-convex constraints. Finally, numerical results validate our analysis and reveal that our proposed multi-static ISAC scheme achieves better ISAC performance than the conventional mono-static ones when the number of cooperative REs is large.