3D Gaussian Splatting aided Localization for Large and Complex Indoor-Environments

The field of visual localization has been researched for several decades and has meanwhile found many practical applications. Despite the strong progress in this field, there are still challenging situations in which established methods fail. We present an approach to significantly improve the accuracy and reliability of established visual localization methods by adding rendered images. In detail, we first use a modern visual SLAM approach that provides a 3D Gaussian Splatting (3DGS) based map to create reference data. We demonstrate that enriching reference data with images rendered from 3DGS at randomly sampled poses significantly improves the performance of both geometry-based visual localization and Scene Coordinate Regression (SCR) methods. Through comprehensive evaluation in a large industrial environment, we analyze the performance impact of incorporating these additional rendered views.

SimpleVQA: Multimodal Factuality Evaluation for Multimodal Large Language Models

The increasing application of multi-modal large language models (MLLMs) across various sectors have spotlighted the essence of their output reliability and accuracy, particularly their ability to produce content grounded in factual information (e.g. common and domain-specific knowledge). In this work, we introduce SimpleVQA, the first comprehensive multi-modal benchmark to evaluate the factuality ability of MLLMs to answer natural language short questions. SimpleVQA is characterized by six key features: it covers multiple tasks and multiple scenarios, ensures high quality and challenging queries, maintains static and timeless reference answers, and is straightforward to evaluate. Our approach involves categorizing visual question-answering items into 9 different tasks around objective events or common knowledge and situating these within 9 topics. Rigorous quality control processes are implemented to guarantee high-quality, concise, and clear answers, facilitating evaluation with minimal variance via an LLM-as-a-judge scoring system. Using SimpleVQA, we perform a comprehensive assessment of leading 18 MLLMs and 8 text-only LLMs, delving into their image comprehension and text generation abilities by identifying and analyzing error cases.

Anti-Degeneracy Scheme for Lidar SLAM based on Particle Filter in Geometry Feature-Less Environments

Simultaneous localization and mapping (SLAM) based on particle filtering has been extensively employed in indoor scenarios due to its high efficiency. However, in geometry feature-less scenes, the accuracy is severely reduced due to lack of constraints. In this article, we propose an anti-degeneracy system based on deep learning. Firstly, we design a scale-invariant linear mapping to convert coordinates in continuous space into discrete indexes, in which a data augmentation method based on Gaussian model is proposed to ensure the model performance by effectively mitigating the impact of changes in the number of particles on the feature distribution. Secondly, we develop a degeneracy detection model using residual neural networks (ResNet) and transformer which is able to identify degeneracy by scrutinizing the distribution of the particle population. Thirdly, an adaptive anti-degeneracy strategy is designed, which first performs fusion and perturbation on the resample process to provide rich and accurate initial values for the pose optimization, and use a hierarchical pose optimization combining coarse and fine matching, which is able to adaptively adjust the optimization frequency and the sensor trustworthiness according to the degree of degeneracy, in order to enhance the ability of searching the global optimal pose. Finally, we demonstrate the optimality of the model, as well as the improvement of the image matrix method and GPU on the computation time through ablation experiments, and verify the performance of the anti-degeneracy system in different scenarios through simulation experiments and real experiments. This work has been submitted to IEEE for publication. Copyright may be transferred without notice, after which this version may no longer be available.

Continuous-Aperture Array Based OAM High-Capacity Communication For Metaverse

The extensive data interaction demands of an immersive metaverse necessitate the adoption of emerging technologies to enable high-capacity communication. Vortex electromagnetic waves with different orbital angular momentum (OAM) modes are spatially orthogonal, providing a novel spatial multiplexing dimension to achieve high-capacity communication. However, the number of orthogonal OAM modes based on a discrete uniform circular array (UCA) is limited by the number of array elements in the UCA, and traditional discrete channel models are unable to accurately capture the physical properties of vortex electromagnetic wave propagation. The continuous-aperture array (CAPA) is composed of densely packed electromagnetic excitation elements, capable of flexibly and efficiently generating the desired surface currents to produce an arbitrary number of mutually orthogonal OAM modes. From the perspective of electromagnetic information theory (EIT), we propose a CAPA-based OAM orthogonal transmission scheme to realize high-capacity communication. We design the surface currents of the CAPA using Fourier basis functions, derive the electromagnetic channel for vortex electromagnetic waves, and investigate the upper bound of the spectrum efficiency for CAPA-based OAM orthogonal transmission. This paper establishes a theoretical foundation for applying EIT to the orthogonal transmission of vortex electromagnetic waves, offering a novel solution for achieving CAPA-based efficient and high-capacity communication.

Collaborative Filtering Meets Spectrum Shift: Connecting User-Item Interaction with Graph-Structured Side Information

Graph Neural Network (GNN) has demonstrated their superiority in collaborative filtering, where the user-item (U-I) interaction bipartite graph serves as the fundamental data format. However, when graph-structured side information (e.g., multimodal similarity graphs or social networks) is integrated into the U-I bipartite graph, existing graph collaborative filtering methods fall short of achieving satisfactory performance. We quantitatively analyze this problem from a spectral perspective. Recall that a bipartite graph possesses a full spectrum within the range of [-1, 1], with the highest frequency exactly achievable at -1 and the lowest frequency at 1; however, we observe as more side information is incorporated, the highest frequency of the augmented adjacency matrix progressively shifts rightward. This spectrum shift phenomenon has caused previous approaches built for the full spectrum [-1, 1] to assign mismatched importance to different frequencies. To this end, we propose Spectrum Shift Correction (dubbed SSC), incorporating shifting and scaling factors to enable spectral GNNs to adapt to the shifted spectrum. Unlike previous paradigms of leveraging side information, which necessitate tailored designs for diverse data types, SSC directly connects traditional graph collaborative filtering with any graph-structured side information. Experiments on social and multimodal recommendation demonstrate the effectiveness of SSC, achieving relative improvements of up to 23% without incurring any additional computational overhead.

NatureLM: Deciphering the Language of Nature for Scientific Discovery

Foundation models have revolutionized natural language processing and artificial intelligence, significantly enhancing how machines comprehend and generate human languages. Inspired by the success of these foundation models, researchers have developed foundation models for individual scientific domains, including small molecules, materials, proteins, DNA, and RNA. However, these models are typically trained in isolation, lacking the ability to integrate across different scientific domains. Recognizing that entities within these domains can all be represented as sequences, which together form the "language of nature", we introduce Nature Language Model (briefly, NatureLM), a sequence-based science foundation model designed for scientific discovery. Pre-trained with data from multiple scientific domains, NatureLM offers a unified, versatile model that enables various applications including: (i) generating and optimizing small molecules, proteins, RNA, and materials using text instructions; (ii) cross-domain generation/design, such as protein-to-molecule and protein-to-RNA generation; and (iii) achieving state-of-the-art performance in tasks like SMILES-to-IUPAC translation and retrosynthesis on USPTO-50k. NatureLM offers a promising generalist approach for various scientific tasks, including drug discovery (hit generation/optimization, ADMET optimization, synthesis), novel material design, and the development of therapeutic proteins or nucleotides. We have developed NatureLM models in different sizes (1 billion, 8 billion, and 46.7 billion parameters) and observed a clear improvement in performance as the model size increases.

Multi-Agent Collaboration for Multilingual Code Instruction Tuning

Recent advancement in code understanding and generation demonstrates that code LLMs fine-tuned on a high-quality instruction dataset can gain powerful capabilities to address wide-ranging code-related tasks. However, most previous existing methods mainly view each programming language in isolation and ignore the knowledge transfer among different programming languages. To bridge the gap among different programming languages, we introduce a novel multi-agent collaboration framework to enhance multilingual instruction tuning for code LLMs, where multiple language-specific intelligent agent components with generation memory work together to transfer knowledge from one language to another efficiently and effectively. Specifically, we first generate the language-specific instruction data from the code snippets and then provide the generated data as the seed data for language-specific agents. Multiple language-specific agents discuss and collaborate to formulate a new instruction and its corresponding solution (A new programming language or existing programming language), To further encourage the cross-lingual transfer, each agent stores its generation history as memory and then summarizes its merits and faults. Finally, the high-quality multilingual instruction data is used to encourage knowledge transfer among different programming languages to train Qwen2.5-xCoder. Experimental results on multilingual programming benchmarks demonstrate the superior performance of Qwen2.5-xCoder in sharing common knowledge, highlighting its potential to reduce the cross-lingual gap.

DobLIX: A Dual-Objective Learned Index for Log-Structured Merge Trees

In this paper, we introduce DobLIX, a dual-objective learned index specifically designed for Log-Structured Merge(LSM) tree-based key-value stores. Although traditional learned indexes focus exclusively on optimizing index lookups, they often overlook the impact of data access from storage, resulting in performance bottlenecks. DobLIX addresses this by incorporating a second objective, data access optimization, into the learned index training process. This dual-objective approach ensures that both index lookup efficiency and data access costs are minimized, leading to significant improvements in read performance while maintaining write efficiency in real-world LSM-tree systems. Additionally, DobLIX features a reinforcement learning agent that dynamically tunes the system parameters, allowing it to adapt to varying workloads in real-time. Experimental results using real-world datasets demonstrate that DobLIX reduces indexing overhead and improves throughput by 1.19 to 2.21 times compared to state-of-the-art methods within RocksDB, a widely used LSM-tree-based storage engine.

HetSSNet: Spatial-Spectral Heterogeneous Graph Learning Network for Panchromatic and Multispectral Images Fusion

Remote sensing pansharpening aims to reconstruct spatial-spectral properties during the fusion of panchromatic (PAN) images and low-resolution multi-spectral (LR-MS) images, finally generating the high-resolution multi-spectral (HR-MS) images. In the mainstream modeling strategies, i.e., CNN and Transformer, the input images are treated as the equal-sized grid of pixels in the Euclidean space. They have limitations in facing remote sensing images with irregular ground objects. Graph is the more flexible structure, however, there are two major challenges when modeling spatial-spectral properties with graph: \emph{1) constructing the customized graph structure for spatial-spectral relationship priors}; \emph{2) learning the unified spatial-spectral representation through the graph}. To address these challenges, we propose the spatial-spectral heterogeneous graph learning network, named \textbf{HetSSNet}. Specifically, HetSSNet initially constructs the heterogeneous graph structure for pansharpening, which explicitly describes pansharpening-specific relationships. Subsequently, the basic relationship pattern generation module is designed to extract the multiple relationship patterns from the heterogeneous graph. Finally, relationship pattern aggregation module is exploited to collaboratively learn unified spatial-spectral representation across different relationships among nodes with adaptive importance learning from local and global perspectives. Extensive experiments demonstrate the significant superiority and generalization of HetSSNet.

FreqPrior: Improving Video Diffusion Models with Frequency Filtering Gaussian Noise

Text-driven video generation has advanced significantly due to developments in diffusion models. Beyond the training and sampling phases, recent studies have investigated noise priors of diffusion models, as improved noise priors yield better generation results. One recent approach employs the Fourier transform to manipulate noise, marking the initial exploration of frequency operations in this context. However, it often generates videos that lack motion dynamics and imaging details. In this work, we provide a comprehensive theoretical analysis of the variance decay issue present in existing methods, contributing to the loss of details and motion dynamics. Recognizing the critical impact of noise distribution on generation quality, we introduce FreqPrior, a novel noise initialization strategy that refines noise in the frequency domain. Our method features a novel filtering technique designed to address different frequency signals while maintaining the noise prior distribution that closely approximates a standard Gaussian distribution. Additionally, we propose a partial sampling process by perturbing the latent at an intermediate timestep during finding the noise prior, significantly reducing inference time without compromising quality. Extensive experiments on VBench demonstrate that our method achieves the highest scores in both quality and semantic assessments, resulting in the best overall total score. These results highlight the superiority of our proposed noise prior.