arXiv subjects

Wei Li

Publications and source records attributed to Wei Li.

At least 451 records · Page 25Linked to original sources

arXiv2025

Joint Optimization of Controller Placement and Switch Assignment in SDN-based LEO Satellite Networks

Software-defined networking (SDN) based low earth orbit (LEO) satellite networks leverage the SDN's benefits of the separation of data plane and control plane, control plane programmability, and centralized control to alleviate the problem of inefficient resource management under traditional network architectures. The most fundamental issue in SDN-based LEO satellite networks is how to place controllers and assign switches. Their outcome directly affects the performance of the network. However, most existing strategies can not sensibly and dynamically adjust the controller location and controller-switch mapping according to the topology variation and traffic undulation of the LEO satellite network meanwhile. In this paper, based on the dynamic placement dynamic assignment scheme, we first formulate the controller placement and switch assignment (CPSA) problem in the LEO satellite networks, which is an integer nonlinear programming problem. Then, a prior population-based genetic algorithm is proposed to solve it. Some individuals of the final generation of the algorithm for the current time slot are used as the prior population of the next time slot, thus stringing together the algorithms of adjacent time slots for successive optimization. Finally, we obtain the near-optimal solution for each time slot. Extensive experiments demonstrate that our algorithm can adapt to the network topology changes and traffic surges, and outperform some existing CPSA strategies in the LEO satellite networks.

cs.NI↗

arXiv2025

Falcon: Advancing Asynchronous BFT Consensus for Lower Latency and Enhanced Throughput

Asynchronous Byzantine Fault Tolerant (BFT) consensus protocols have garnered significant attention with the rise of blockchain technology. A typical asynchronous protocol is designed by executing sequential instances of the Asynchronous Common Sub-seQuence (ACSQ). The ACSQ protocol consists of two primary components: the Asynchronous Common Subset (ACS) protocol and a block sorting mechanism, with the ACS protocol comprising two stages: broadcast and agreement. However, current protocols encounter three critical issues: high latency arising from the execution of the agreement stage, latency instability due to the integral-sorting mechanism, and reduced throughput caused by block discarding. To address these issues,we propose Falcon, an asynchronous BFT protocol that achieves low latency and enhanced throughput. Falcon introduces a novel broadcast protocol, Graded Broadcast (GBC), which enables a block to be included in the ACS set directly, bypassing the agreement stage and thereby reducing latency. To ensure safety, Falcon incorporates a new binary agreement protocol called Asymmetrical Asynchronous Binary Agreement (AABA), designed to complement GBC. Additionally, Falcon employs a partial-sorting mechanism, allowing continuous rather than simultaneous block committing, enhancing latency stability. Finally, we incorporate an agreement trigger that, before its activation, enables nodes to wait for more blocks to be delivered and committed, thereby boosting throughput. We conduct a series of experiments to evaluate Falcon, demonstrating its superior performance.

cs.DC↗

arXiv2025

PSBD: Prediction Shift Uncertainty Unlocks Backdoor Detection

Deep neural networks are susceptible to backdoor attacks, where adversaries manipulate model predictions by inserting malicious samples into the training data. Currently, there is still a significant challenge in identifying suspicious training data to unveil potential backdoor samples. In this paper, we propose a novel method, Prediction Shift Backdoor Detection (PSBD), leveraging an uncertainty-based approach requiring minimal unlabeled clean validation data. PSBD is motivated by an intriguing Prediction Shift (PS) phenomenon, where poisoned models' predictions on clean data often shift away from true labels towards certain other labels with dropout applied during inference, while backdoor samples exhibit less PS. We hypothesize PS results from the neuron bias effect, making neurons favor features of certain classes. PSBD identifies backdoor training samples by computing the Prediction Shift Uncertainty (PSU), the variance in probability values when dropout layers are toggled on and off during model inference. Extensive experiments have been conducted to verify the effectiveness and efficiency of PSBD, which achieves state-of-the-art results among mainstream detection methods. The code is available at https://github.com/WL-619/PSBD.

cs.LG↗

arXiv2025

Semiparametric Causal Discovery and Inference with Invalid Instruments

Learning causal relationships among a set of variables, as encoded by a directed acyclic graph, from observational data is complicated by the presence of unobserved confounders. Instrumental variables (IVs) are a popular remedy for this issue, but most existing methods either assume the validity of all IVs or postulate a specific form of relationship, such as a linear model, between the primary variables and the IVs. To overcome these limitations, we introduce a partially linear structural equation model for causal discovery and inference that accommodates potentially invalid IVs and allows for general dependence of the primary variables on the IVs. We establish identification under this semiparametric model by constructing surrogate valid IVs, and develop a finite-sample procedure for estimating the causal structures and effects. Theoretically, we show that our procedure consistently learns the causal structures, yields asymptotically normal estimates, and effectively controls the false discovery rate in edge recovery. Simulation studies demonstrate the superiority of our method over existing competitors, and an application to inferring gene regulatory networks in Alzheimer's disease illustrates its usefulness.

stat.ME↗

arXiv2025

Giant Magnetocaloric Effect in Spin Supersolid Candidate Na$_2$BaCo(PO$_4$)$_2$

Supersolid, an exotic quantum state of matter that consists of particles forming an incompressible solid structure while simultaneously showing superfluidity of zero viscosity [1], is one of the long-standing pursuits in fundamental research [2, 3]. Although the initial report of $^4$He supersolid turned out to be an artifact [4], this intriguing quantum matter has inspired enthusiastic investigations into ultracold quantum gases [5-8]. Nevertheless, the realization of supersolidity in condensed matter remains elusive. Here we find evidence for a quantum magnetic analogue of supersolid -- the spin supersolid -- in the recently synthesized triangular-lattice antiferromagnet Na$_2$BaCo(PO$_4$)$_2$ [9]. Notably, a giant magnetocaloric effect related to the spin supersolidity is observed in the demagnetization cooling process, manifesting itself as two prominent valley-like regimes, with the lowest temperature attaining below 100 mK. Not only is there an experimentally determined series of critical fields but the demagnetization cooling profile also shows excellent agreement with the theoretical simulations with an easy-axis Heisenberg model. Neutron diffractions also successfully locate the proposed spin supersolid phases by revealing the coexistence of three-sublattice spin solid order and interlayer incommensurability indicative of the spin superfluidity. Thus, our results indicate a strong entropic effect of the spin supersolid phase in a frustrated quantum magnet and open up a viable and promising avenue for applications in sub-Kelvin refrigeration, especially in the context of persistent concerns about helium shortages [10, 11].

cond-mat.str-el↗

arXiv2025

Ultra-Low Complexity On-Orbit Compression for Remote Sensing Imagery via Block Modulated Imaging

The growing field of remote sensing faces a challenge: the ever-increasing size and volume of imagery data are exceeding the storage and transmission capabilities of satellite platforms. Efficient compression of remote sensing imagery is a critical solution to alleviate these burdens on satellites. However, existing compression methods are often too computationally expensive for satellites. With the continued advancement of compressed sensing theory, single-pixel imaging emerges as a powerful tool that brings new possibilities for on-orbit image compression. However, it still suffers from prolonged imaging times and the inability to perform high-resolution imaging, hindering its practical application. This paper advances the study of compressed sensing in remote sensing image compression, proposing Block Modulated Imaging (BMI). By requiring only a single exposure, BMI significantly enhances imaging acquisition speeds. Additionally, BMI obviates the need for digital micromirror devices and surpasses limitations in image resolution. Furthermore, we propose a novel decoding network specifically designed to reconstruct images compressed under the BMI framework. Leveraging the gated 3D convolutions and promoting efficient information flow across stages through a Two-Way Cross-Attention module, our decoding network exhibits demonstrably superior reconstruction performance. Extensive experiments conducted on multiple renowned remote sensing datasets unequivocally demonstrate the efficacy of our proposed method. To further validate its practical applicability, we developed and tested a prototype of the BMI-based camera, which has shown promising potential for on-orbit image compression. The code is available at https://github.com/Johnathan218/BMNet.

eess.IV↗

arXiv2025

Low latency global carbon budget reveals a continuous decline of the land carbon sink during the 2023/24 El Nino event

The high growth rate of atmospheric CO2 in 2023 was found to be caused by a severe reduction of the global net land carbon sink. Here we update the global CO2 budget from January 1st to July 1st 2024, during which El Niño drought conditions continued to prevail in the Tropics but ceased by March 2024. We used three dynamic global vegetation models (DGVMs), machine learning emulators of ocean models, three atmospheric inversions driven by observations from the second Orbiting Carbon Observatory (OCO-2) satellite, and near-real-time fossil CO2 emissions estimates. In a one-year period from July 2023 to July 2024 covering the El Niño 2023/24 event, we found a record-high CO2 growth rate of 3.66~$\pm$~0.09 ppm~yr$^{-1}$ ($\pm$~1 standard deviation) since 1979. Yet, the CO2 growth rate anomaly obtained after removing the long term trend is 1.1 ppm~yr$^{-1}$, which is marginally smaller than the July--July growth rate anomalies of the two major previous El Niño events in 1997/98 and 2015/16. The atmospheric CO2 growth rate anomaly was primarily driven by a 2.24 GtC~yr$^{-1}$ reduction in the net land sink including 0.3 GtC~yr$^{-1}$ of fire emissions, partly offset by a 0.38 GtC~yr$^{-1}$ increase in the ocean sink relative to the 2015--2022 July--July mean. The tropics accounted for 97.5\% of the land CO2 flux anomaly, led by the Amazon (50.6\%), central Africa (34\%), and Southeast Asia (8.2\%), with extra-tropical sources in South Africa and southern Brazil during April--July 2024. Our three DGVMs suggest greater tropical CO2 losses in 2023/2024 than during the two previous large El Niño in 1997/98 and 2015/16, whereas inversions indicate losses more comparable to 2015/16. Overall, this update of the low latency budget highlights the impact of recent El Niño droughts in explaining the high CO2 growth rate until July 2024.

physics.ao-ph↗

arXiv2025

Spiking Neural Network for Intra-cortical Brain Signal Decoding

Decoding brain signals accurately and efficiently is crucial for intra-cortical brain-computer interfaces. Traditional decoding approaches based on neural activity vector features suffer from low accuracy, whereas deep learning based approaches have high computational cost. To improve both the decoding accuracy and efficiency, this paper proposes a spiking neural network (SNN) for effective and energy-efficient intra-cortical brain signal decoding. We also propose a feature fusion approach, which integrates the manually extracted neural activity vector features with those extracted by a deep neural network, to further improve the decoding accuracy. Experiments in decoding motor-related intra-cortical brain signals of two rhesus macaques demonstrated that our SNN model achieved higher accuracy than traditional artificial neural networks; more importantly, it was tens or hundreds of times more efficient. The SNN model is very suitable for high precision and low power applications like intra-cortical brain-computer interfaces.

cs.HC↗

arXiv2025

F-LMM: Grounding Frozen Large Multimodal Models

Endowing Large Multimodal Models (LMMs) with visual grounding capability can significantly enhance AIs' understanding of the visual world and their interaction with humans. However, existing methods typically fine-tune the parameters of LMMs to learn additional segmentation tokens and overfit grounding and segmentation datasets. Such a design would inevitably cause a catastrophic diminution in the indispensable conversational capability of general AI assistants. In this paper, we comprehensively evaluate state-of-the-art grounding LMMs across a suite of multimodal question-answering benchmarks, observing drastic performance drops that indicate vanishing general knowledge comprehension and weakened instruction following ability. To address this issue, we present F-LMM -- grounding frozen off-the-shelf LMMs in human-AI conversations -- a straightforward yet effective design based on the fact that word-pixel correspondences conducive to visual grounding inherently exist in the attention mechanism of well-trained LMMs. Using only a few trainable CNN layers, we can translate word-pixel attention weights to mask logits, which a SAM-based mask refiner can further optimise. Our F-LMM neither learns special segmentation tokens nor utilises high-quality grounded instruction-tuning data, but achieves competitive performance on referring expression segmentation and panoptic narrative grounding benchmarks while completely preserving LMMs' original conversational ability. Additionally, with instruction-following ability preserved and grounding ability obtained, F-LMM can be directly applied to complex tasks like reasoning segmentation, grounded conversation generation and visual chain-of-thought reasoning. Our code can be found at https://github.com/wusize/F-LMM.

cs.CV↗

arXiv2025

Pangu Ultra: Pushing the Limits of Dense Large Language Models on Ascend NPUs

We present Pangu Ultra, a Large Language Model (LLM) with 135 billion parameters and dense Transformer modules trained on Ascend Neural Processing Units (NPUs). Although the field of LLM has been witnessing unprecedented advances in pushing the scale and capability of LLM in recent years, training such a large-scale model still involves significant optimization and system challenges. To stabilize the training process, we propose depth-scaled sandwich normalization, which effectively eliminates loss spikes during the training process of deep models. We pre-train our model on 13.2 trillion diverse and high-quality tokens and further enhance its reasoning capabilities during post-training. To perform such large-scale training efficiently, we utilize 8,192 Ascend NPUs with a series of system optimizations. Evaluations on multiple diverse benchmarks indicate that Pangu Ultra significantly advances the state-of-the-art capabilities of dense LLMs such as Llama 405B and Mistral Large 2, and even achieves competitive results with DeepSeek-R1, whose sparse model structure contains much more parameters. Our exploration demonstrates that Ascend NPUs are capable of efficiently and effectively training dense models with more than 100 billion parameters. Our model and system will be available for our commercial customers.

cs.CL↗

arXiv2025

Respiratory Differencing: Enhancing Pulmonary Thermal Ablation Evaluation Through Pre- and Intra-Operative Image Fusion

CT image-guided thermal ablation is widely used for lung cancer treatment; however, follow-up data indicate that physicians' subjective assessments of intraoperative images often overestimate the ablation effect, potentially leading to incomplete treatment. To address these challenges, we developed \textit{Respiratory Differencing}, a novel intraoperative CT image assistance system aimed at improving ablation evaluation. The system first segments tumor regions in preoperative CT images and then employs a multi-stage registration process to align these images with corresponding intraoperative or postoperative images, compensating for respiratory deformations and treatment-induced changes. This system provides two key outputs to help physicians evaluate intraoperative ablation. First, differential images are generated by subtracting the registered preoperative images from the intraoperative ones, allowing direct visualization and quantitative comparison of pre- and post-treatment differences. These differential images enable physicians to assess the relative positions of the tumor and ablation zones, even when the tumor is no longer visible in post-ablation images, thus improving the subjective evaluation of ablation effectiveness. Second, the system provides a quantitative metric that measures the discrepancies between the tumor area and the treatment zone, offering a numerical assessment of the overall efficacy of ablation.This pioneering system compensates for complex lung deformations and integrates pre- and intra-operative imaging data, enhancing quality control in cancer ablation treatments. A follow-up study involving 35 clinical cases demonstrated that our system significantly outperforms traditional subjective assessments in identifying under-ablation cases during or immediately after treatment, highlighting its potential to improve clinical decision-making and patient outcomes.

cs.CV↗

arXiv2025

BRIDGES: Bridging Graph Modality and Large Language Models within EDA Tasks

While many EDA tasks already involve graph-based data, existing LLMs in EDA primarily either represent graphs as sequential text, or simply ignore graph-structured data that might be beneficial like dataflow graphs of RTL code. Recent studies have found that LLM performance suffers when graphs are represented as sequential text, and using additional graph information significantly boosts performance. To address these challenges, we introduce BRIDGES, a framework designed to incorporate graph modality into LLMs for EDA tasks. BRIDGES integrates an automated data generation workflow, a solution that combines graph modality with LLM, and a comprehensive evaluation suite. First, we establish an LLM-driven workflow to generate RTL and netlist-level data, converting them into dataflow and netlist graphs with function descriptions. This workflow yields a large-scale dataset comprising over 500,000 graph instances and more than 1.5 billion tokens. Second, we propose a lightweight cross-modal projector that encodes graph representations into text-compatible prompts, enabling LLMs to effectively utilize graph data without architectural modifications. Experimental results demonstrate 2x to 10x improvements across multiple tasks compared to text-only baselines, including accuracy in design retrieval, type prediction and perplexity in function description, with negligible computational overhead (<1% model weights increase and <30% additional runtime overhead). Even without additional LLM finetuning, our results outperform text-only by a large margin. We plan to release BRIDGES, including the dataset, models, and training flow.

cs.LG↗

arXiv2025

AndroidWorld: A Dynamic Benchmarking Environment for Autonomous Agents

Autonomous agents that execute human tasks by controlling computers can enhance human productivity and application accessibility. However, progress in this field will be driven by realistic and reproducible benchmarks. We present AndroidWorld, a fully functional Android environment that provides reward signals for 116 programmatic tasks across 20 real-world Android apps. Unlike existing interactive environments, which provide a static test set, AndroidWorld dynamically constructs tasks that are parameterized and expressed in natural language in unlimited ways, thus enabling testing on a much larger and more realistic suite of tasks. To ensure reproducibility, each task includes dedicated initialization, success-checking, and tear-down logic, which modifies and inspects the device's system state. We experiment with baseline agents to test AndroidWorld and provide initial results on the benchmark. Our best agent can complete 30.6% of AndroidWorld's tasks, leaving ample room for future work. Furthermore, we adapt a popular desktop web agent to work on Android, which we find to be less effective on mobile, suggesting future research is needed to achieve universal, cross-platform agents. Finally, we also conduct a robustness analysis, showing that task variations can significantly affect agent performance, demonstrating that without such testing, agent performance metrics may not fully reflect practical challenges. AndroidWorld and the experiments in this paper are available at github.com/google-research/android_world.

cs.AI↗

arXiv2025

GMAI-VL & GMAI-VL-5.5M: A Large Vision-Language Model and A Comprehensive Multimodal Dataset Towards General Medical AI

Despite significant advancements in general AI, its effectiveness in the medical domain is limited by the lack of specialized medical knowledge. To address this, we formulate GMAI-VL-5.5M, a multimodal medical dataset created by converting hundreds of specialized medical datasets with various annotations into high-quality image-text pairs. This dataset offers comprehensive task coverage, diverse modalities, and rich image-text data. Building upon this dataset, we develop GMAI-VL, a general medical vision-language model, with a three-stage training strategy that enhances the integration of visual and textual information. This approach significantly improves the model's ability to process multimodal data, supporting accurate diagnoses and clinical decision-making. Experiments show that GMAI-VL achieves state-of-the-art performance across various multimodal medical tasks, including visual question answering and medical image diagnosis.

cs.CV↗

arXiv2025

A Deep Learning Pipeline for Large Earthquake Analysis using High-Rate Global Navigation Satellite System Data

Deep learning techniques for processing large and complex datasets have unlocked new opportunities for fast and reliable earthquake analysis using Global Navigation Satellite System (GNSS) data. This work presents a deep learning model, MagEs, to estimate earthquake magnitudes using data from high-rate GNSS stations. Furthermore, MagEs is integrated with the DetEQ model for earthquake detection within the SAIPy package, creating a comprehensive pipeline for earthquake detection and magnitude estimation using HR-GNSS data. The MagEs model provides magnitude estimates within seconds of detection when using stations within 3 degrees of the epicenter, which are the most relevant for real-time applications. However, since it has been trained on data from stations up to 7.5 degrees away, it can also analyze data from larger distances. The model can process data from a single station at a time or combine data from up to three stations. The model was trained using synthetic data reflecting rupture scenarios in the Chile subduction zone, and the results confirm strong performance for Chilean earthquakes. Although tests from other tectonic regions also yielded good results, incorporating regional data through transfer learning could further improve its performance in diverse seismic settings. The model has not yet been deployed in an operational real-time monitoring system, but simulation tests that update data in a second-by-second manner demonstrate its potential for future real-time adaptation.

physics.geo-ph↗

arXiv2025

Free4D: Tuning-free 4D Scene Generation with Spatial-Temporal Consistency

We present Free4D, a novel tuning-free framework for 4D scene generation from a single image. Existing methods either focus on object-level generation, making scene-level generation infeasible, or rely on large-scale multi-view video datasets for expensive training, with limited generalization ability due to the scarcity of 4D scene data. In contrast, our key insight is to distill pre-trained foundation models for consistent 4D scene representation, which offers promising advantages such as efficiency and generalizability. 1) To achieve this, we first animate the input image using image-to-video diffusion models followed by 4D geometric structure initialization. 2) To turn this coarse structure into spatial-temporal consistent multiview videos, we design an adaptive guidance mechanism with a point-guided denoising strategy for spatial consistency and a novel latent replacement strategy for temporal coherence. 3) To lift these generated observations into consistent 4D representation, we propose a modulation-based refinement to mitigate inconsistencies while fully leveraging the generated information. The resulting 4D representation enables real-time, controllable rendering, marking a significant advancement in single-image-based 4D scene generation.

cs.CV↗

arXiv2025

Purely Speckled Intensity Images Need for SAR Despeckling with SDS-SAR

Speckle noise is generated along with the SAR imaging mechanism and degrades the quality of SAR images, leading to difficult interpretation. Hence, despeckling is an indispensable step in SAR pre-processing. Fortunately, supervised learning (SL) has proven to be a progressive method for SAR image despeckling. SL methods necessitate the availability of both original SAR images and their speckle-free counterparts during training, whilst speckle-free SAR images do not exist in the real world. Even though there are several substitutes for speckle-free images, the domain gap leads to poor performance and adaptability. Self-supervision provides an approach to training without clean reference. However, most self-supervised methods impose high demands on speckle modeling or specific data, limiting their practicality in real-world applications. To address these challenges, we propose a Self-supervised Despeckling Strategy for SAR images (SDS-SAR) that relies solely on speckled intensity data for training. Firstly, the theoretical feasibility of SAR image despeckling without speckle-free images is established. A self-supervised despeckling criteria suitable for all SAR images is proposed. Subsequently, a Random-Aware sub-SAMpler with Projection correLation Estimation (RA-SAMPLE) is put forth. Mutually independent training pairs can be derived from actual SAR intensity images. Furthermore, a multi-feature loss function is introduced, consisting of a despeckling term, a regularization term, and a perception term. The performance of speckle suppression and texture preservation is well-balanced. Experiments reveal that the proposed method performs on par with supervised approaches on synthetic data and outperforms them on actual data.

eess.IV↗

arXiv2025

ISPDiffuser: Learning RAW-to-sRGB Mappings with Texture-Aware Diffusion Models and Histogram-Guided Color Consistency

RAW-to-sRGB mapping, or the simulation of the traditional camera image signal processor (ISP), aims to generate DSLR-quality sRGB images from raw data captured by smartphone sensors. Despite achieving comparable results to sophisticated handcrafted camera ISP solutions, existing learning-based methods still struggle with detail disparity and color distortion. In this paper, we present ISPDiffuser, a diffusion-based decoupled framework that separates the RAW-to-sRGB mapping into detail reconstruction in grayscale space and color consistency mapping from grayscale to sRGB. Specifically, we propose a texture-aware diffusion model that leverages the generative ability of diffusion models to focus on local detail recovery, in which a texture enrichment loss is further proposed to prompt the diffusion model to generate more intricate texture details. Subsequently, we introduce a histogram-guided color consistency module that utilizes color histogram as guidance to learn precise color information for grayscale to sRGB color consistency mapping, with a color consistency loss designed to constrain the learned color information. Extensive experimental results show that the proposed ISPDiffuser outperforms state-of-the-art competitors both quantitatively and visually. The code is available at https://github.com/RenYangSCU/ISPDiffuser.

cs.CV↗