arXiv subjects

Liang Liu

Publications and source records attributed to Liang Liu.

At least 73 records · Page 4Linked to original sources

arXiv2025

VLM-Assisted Continual learning for Visual Question Answering in Self-Driving

In this paper, we propose a novel approach for solving the Visual Question Answering (VQA) task in autonomous driving by integrating Vision-Language Models (VLMs) with continual learning. In autonomous driving, VQA plays a vital role in enabling the system to understand and reason about its surroundings. However, traditional models often struggle with catastrophic forgetting when sequentially exposed to new driving tasks, such as perception, prediction, and planning, each requiring different forms of knowledge. To address this challenge, we present a novel continual learning framework that combines VLMs with selective memory replay and knowledge distillation, reinforced by task-specific projection layer regularization. The knowledge distillation allows a previously trained model to act as a "teacher" to guide the model through subsequent tasks, minimizing forgetting. Meanwhile, task-specific projection layers calculate the loss based on the divergence of feature representations, ensuring continuity in learning and reducing the shift between tasks. Evaluated on the DriveLM dataset, our framework shows substantial performance improvements, with gains ranging from 20.11% to 35.16% across various metrics. These results highlight the effectiveness of combining continual learning with VLMs in enhancing the resilience and reliability of VQA systems in autonomous driving. We will release our source code.

cs.CV↗

arXiv2025

Fast and Accurate Node-Age Estimation Under Fossil Calibration Uncertainty Using the Adjusted Pairwise Likelihood

Estimating divergence times from molecular sequence data is central to reconstructing the evolutionary history of lineages. Although Bayesian relaxed-clock methods provide a principled framework for incorporating fossil information, their dependence on repeated evaluations of the full phylogenetic likelihood makes them computationally demanding for large genomic datasets. Furthermore, because disagreements in divergence-time estimates often arise from uncertainty or error in fossil placement and prior specification, there is a need for methods that are both computationally efficient and robust to fossil-calibration uncertainty. In this study, we introduce fast and accurate alternatives based on the phylogenetic pairwise composite likelihood, presenting two adjusted pairwise likelihood (APW) formulations that employ asymptotic moment-matching weights to better approximate the behavior of the full likelihood within a Bayesian MCMC framework. Extensive simulations across diverse fossil-calibration scenarios show that APW methods produce node-age estimates comparable to those obtained from the full likelihood while offering greater robustness to fossil misplacement and prior misspecification, due to the reduced sensitivity of composite likelihoods to local calibration errors. Applied to a genome-scale dataset of modern birds, APW methods recover divergence time patterns consistent with recent studies, while reducing computational cost by more than an order of magnitude. Overall, our results demonstrate that adjusted pairwise likelihoods provide a calibration-robust and computationally efficient framework for Bayesian node dating, especially suited for large phylogenomic datasets and analyses in which fossil priors may be uncertain or imperfectly placed.

q-bio.QM↗

arXiv2025

Improving Partially Observed Trajectories Forecasting by Target-driven Self-Distillation

Accurate prediction of future trajectories of traffic agents is essential for ensuring safe autonomous driving. However, partially observed trajectories can significantly degrade the performance of even state-of-the-art models. Previous approaches often rely on knowledge distillation to transfer features from fully observed trajectories to partially observed ones. This involves firstly training a fully observed model and then using a distillation process to create the final model. While effective, they require multi-stage training, making the training process very expensive. Moreover, knowledge distillation can lead to a performance degradation of the model. In this paper, we introduce a Target-drivenSelf-Distillation method (TSD) for motion forecasting. Our method leverages predicted accurate targets to guide the model in making predictions under partial observation conditions. By employing self-distillation, the model learns from the feature distributions of both fully observed and partially observed trajectories during a single end-to-end training process. This enhances the model's ability to predict motion accurately in both fully observed and partially observed scenarios. We evaluate our method on multiple datasets and state-of-the-art motion forecasting models. Extensive experimental results demonstrate that our approach achieves significant performance improvements in both settings. To facilitate further research, we will release our code and model checkpoints.

cs.CV↗

arXiv2025

RoboDriveVLM: A Novel Benchmark and Baseline towards Robust Vision-Language Models for Autonomous Driving

Current Vision-Language Model (VLM)-based end-to-end autonomous driving systems often leverage large language models to generate driving decisions directly based on their understanding of the current scene. However, such systems introduce multiple risks in real-world driving scenarios. To evaluate whether VLMs are truly viable for autonomous driving, we introduce RoboDriveBench, the first robustness benchmark focused on end-to-end trajectory prediction tasks. This benchmark systematically evaluates two critical categories of real-world challenges for VLM-based end-to-end autonomous driving systems through 11 simulated scenarios encompassing various corruption types, including 6 scenarios of sensor corruption caused by environmental variations, along with 5 cases of prompt corruption resulting from human intervention and data transmission failures. Each corruption type includes 250 unique driving scenarios and 5,689 frames, resulting in 64,559 total trajectory prediction cases per evaluation. To overcome these real-world challenges, we propose a novel VLM-based autonomous driving framework called RoboDriveVLM, which enhances robustness by mapping more multimodal data-e.g., lidar and radar-into a unified latent space. Furthermore, we introduce a new Test-Time Adaptation (TTA) method based on cross-modal knowledge distillation to improve the robustness of VLM-based autonomous driving systems. Through extensive experiments, our work highlights the limitations of current VLM-based end-to-end autonomous driving systems and provides a more reliable solution for real-world deployment. Source code and datasets will be released.

cs.AI↗

arXiv2025

Federated Semi-Supervised and Semi-Asynchronous Learning for Anomaly Detection in IoT Networks

Existing FL-based approaches are based on the unrealistic assumption that the data on the client-side is fully annotated with ground truths. Furthermore, it is a great challenge how to improve the training efficiency while ensuring the detection accuracy in the highly heterogeneous and resource-constrained IoT networks. Meanwhile, the communication cost between clients and the server is also a problem that can not be ignored. Therefore, in this paper, we propose a Federated Semi-Supervised and Semi-Asynchronous (FedS3A) learning for anomaly detection in IoT networks. First, we consider a more realistic assumption that labeled data is only available at the server, and pseudo-labeling is utilized to implement federated semi-supervised learning, in which a dynamic weight of supervised learning is exploited to balance the supervised learning at the server and unsupervised learning at clients. Then, we propose a semi-asynchronous model update and staleness tolerant distribution scheme to achieve a trade-off between the round efficiency and detection accuracy. Meanwhile, the staleness of local models and the participation frequency of clients are considered to adjust their contributions to the global model. In addition, a group-based aggregation function is proposed to deal with the non-IID distribution of the data. Finally, the difference transmission based on the sparse matrix is adopted to reduce the communication cost. Extensive experimental results show that FedS3A can achieve greater than 98% accuracy even when the data is non-IID and is superior to the classic FL-based algorithms in terms of both detection performance and round efficiency, achieving a win-win situation. Meanwhile, FedS3A successfully reduces the communication cost by higher than 50%.

cs.DC↗

arXiv2025

Coexistence of unconventional spin-orbit torque and in-plane Hall effect in a single ferromagnetic layer

The symmetry of a material fundamentally governs its spin transport properties. While unconventional spin transport phenomena have been predominantly explored in low-symmetry systems (e.g., $C_{1v}$ symmetry), high-symmetry crystals--which constitute the majority of industry-compatible materials--are generally expected to exhibit only conventional spin-transport behavior. Here, we report the coexistence of two unconventional spin transport effects, the crystal spin-orbit torque (CSOT) and the crystal in-plane Hall effect (CIHE), in a CoPt single ferromagnetic layer with $C_{3v}$ symmetry. Leveraging the CSOT, we achieve nearly 100% field-free perpendicular magnetization switching in a 6 nm CoPt layer at room temperature. Simultaneously, the CIHE observed in this material exhibits nearly identical dependencies on both current angle and growth temperature as the CSOT. Symmetry analysis confirms that both effects share a common physical origin. Our work not only establishes CoPt as a high-performance spin-orbit material, but also demonstrates that unconventional spin transport can be realized in high-symmetry systems, thereby opening a broad pathway for their application in practical spintronics.

cond-mat.mes-hall↗

arXiv2025

LLM-Powered GUI Agents in Phone Automation: Surveying Progress and Prospects

With the rapid rise of large language models (LLMs), phone automation has undergone transformative changes. This paper systematically reviews LLM-driven phone GUI agents, highlighting their evolution from script-based automation to intelligent, adaptive systems. We first contextualize key challenges, (i) limited generality, (ii) high maintenance overhead, and (iii) weak intent comprehension, and show how LLMs address these issues through advanced language understanding, multimodal perception, and robust decision-making. We then propose a taxonomy covering fundamental agent frameworks (single-agent, multi-agent, plan-then-act), modeling approaches (prompt engineering, training-based), and essential datasets and benchmarks. Furthermore, we detail task-specific architectures, supervised fine-tuning, and reinforcement learning strategies that bridge user intent and GUI operations. Finally, we discuss open challenges such as dataset diversity, on-device deployment efficiency, user-centric adaptation, and security concerns, offering forward-looking insights into this rapidly evolving field. By providing a structured overview and identifying pressing research gaps, this paper serves as a definitive reference for researchers and practitioners seeking to harness LLMs in designing scalable, user-friendly phone GUI agents. The collection of papers reviewed in this survey will be hosted and regularly updated on the GitHub repository: https://github.com/PhoneLLM/Awesome-LLM-Powered-Phone-GUI-Agents

cs.HC↗

arXiv2025

Stacking-orientation and twist-angle control on integer and fractional Chern insulators in moiré rhombohedral graphene

Rhombohedral-stacked multilayer graphene aligned with hexagonal boron nitride has emerged as an excellent platform for investigating exotic quantum phenomena arising from the interplay between electron correlations and nontrivial topology. However, the microscopic mechanism governing the emergence of both the integer and fractional Chern insulator states in this system remains an open question. In this work, we systematically investigate the electrical transport properties of RMG/hBN moiré devices with controlled alignment orientations and twist angles. We demonstrate that alignment orientation strongly modulates correlated phenomena in the moiré-proximal regime, while having negligible influence on the formation of integer and fractional Chern insulators in the moiré-distant regime. Instead, the moiré periodicity, tuned by the twist angle, serves as the key parameter controlling the stability of these correlated topological states in the moiré-distant regime. Furthermore, in the moiré-proximal regime of one specific alignment, we observe anomalous Hall effect and a variety of competing phases near ν = 1, including integer Chern insulator states, extended Chern insulator states, and trivial insulators, whose stability is highly sensitive to both the applied displacement electric field and magnetic field. Our results underscore the critical role of stacking-alignment and twist-angle engineering in exploring novel quantum states based on rhombohedral-stacked multilayer graphene moiré systems.

cond-mat.mes-hall↗

arXiv2025

DeLoad: Demand-Driven Short-Video Preloading with Scalable Watch-Time Estimation

Short video streaming has become a dominant paradigm in digital media, characterized by rapid swiping interactions and diverse media content. A key technical challenge is designing an effective preloading strategy that dynamically selects and prioritizes download tasks from an evolving playlist, balancing Quality of Experience (QoE) and bandwidth efficiency under practical commercial constraints. However, real world analysis reveals critical limitations of existing approaches: (1) insufficient adaptation of download task sizes to dynamic conditions, and (2) watch time prediction models that are difficult to deploy reliably at scale. In this paper, we propose DeLoad, a novel preloading framework that addresses these issues by introducing dynamic task sizing and a practical, multi dimensional watch time estimation method. Additionally, a Deep Reinforcement Learning (DRL) enhanced agent is trained to optimize the download range decisions adaptively. Extensive evaluations conducted on an offline testing platform, leveraging massive real world network data, demonstrate that DeLoad achieves significant improvements in QoE metrics (34.4% to 87.4% gain). Furthermore, after deployment on a large scale commercial short video platform, DeLoad has increased overall user watch time by 0.09% while simultaneously reducing rebuffering events and 3.76% bandwidth consumption.

cs.MM↗

arXiv2025

PIRA: Pan-CDN Intra-video Resource Adaptation for Short Video Streaming

In large scale short video platforms, CDN resource selection plays a critical role in maintaining Quality of Experience (QoE) while controlling escalating traffic costs. To better understand this phenomenon, we conduct in the wild network measurements during video playback in a production short video system. The results reveal that CDNs delivering higher average QoE often come at greater financial cost, yet their connection quality fluctuates even within a single video underscoring a fundamental and dynamic trade off between QoE and cost. However, the problem of sustaining high QoE under cost constraints remains insufficiently investigated in the context of CDN selection for short video streaming. To address this, we propose PIRA, a dynamic resource selection algorithm that optimizes QoE and cost in real time during video playback. PIRA formally integrating QoE and cost by a mathematical model, and introduce a intra video control theoretic CDN resource selection approach which can balance QoE and cost under network dynamics. To reduce the computation overheads, PIRA employs state space pruning and adaptive parameter adjustment to efficiently solve the high dimensional optimization problem. In large scale production experiments involving 450,000 users over two weeks, PIRA outperforms the production baseline, achieving a 2.1% reduction in start up delay, 15.2% shorter rebuffering time, and 10% lower average unit traffic cost, demonstrating its effectiveness in balancing user experience and financial cost at scale.

cs.MM↗

arXiv2025

Integrated Massive Communication and Target Localization in 6G Cell-Free Networks

This paper presents an initial investigation into the combination of integrated sensing and communication (ISAC) and massive communication, both of which are largely regarded as key scenarios in sixth-generation (6G) wireless networks. Specifically, we consider a cell-free network comprising a large number of users, multiple targets, and distributed base stations (BSs). In each time slot, a random subset of users becomes active, transmitting pilot signals that can be scattered by the targets before reaching the BSs. Unlike conventional massive random access schemes, where the primary objectives are device activity detection and channel estimation, our framework also enables target localization by leveraging the multipath propagation effects introduced by the targets. However, due to the intricate dependency between user channels and target locations, characterizing the posterior distribution required for minimum mean-square error (MMSE) estimation presents significant computational challenges. To handle this problem, we propose a hybrid message passing-based framework that incorporates multiple approximations to mitigate computational complexity. Numerical results demonstrate that the proposed approach achieves high-accuracy device activity detection, channel estimation, and target localization simultaneously, validating the feasibility of embedding localization functionality into massive communication systems for future 6G networks.

eess.SP↗

arXiv2025

Spectrum Efficiency and Processing Latency Trade-offs in Panel-Based LIS

The next generation wireless systems will face stringent new requirements, including ultra-low latency, high data rates and enhanced reliability. Large Intelligent Surfaces, is one proposed solution that has the potential to solve these high demands. The real-life deployment of such systems involves different design considerations with non-trivial trade-offs. This paper investigates the trade-off between spectral efficiency and processing latency, considering different antenna distribution schemes and detection algorithms. A latency model for the physical layer processing has been developed, using real FPGA and application-specific instruction processor (ASIP) hardware implementation results. Simulation results using an indoor environment show that distributing antennas throughout the scenario improves overall reliability, while the impact from this on latency is limited both when using zero-forcing (ZF) and Minimum Mean Square Error (MMSE) detection. Changing the detection algorithm to maximum-ratio combining (MRC) from ZF or MMSE, however, reduces the latency significantly, even if a larger number of antennas are needed to achieve a similar spectrum efficiency.

eess.SP↗

arXiv2025

Highly Parallel Singular Value Decomposition for Low-Latency MIMO Processing

Singular value decomposition (SVD) is widely used in wireless systems, including multiple-input multiple-output (MIMO) processing and dimension reduction in distributed MIMO (D-MIMO). However, the iterative nature of decomposition methods results in increased execution time as system size grows, posing challenges for real-time and low-latency applications. To address this, we analyze the latency of state-of-art SVD methods, and highlight the efficiency of a 4-step highly parallel method based on Gram matrix tridiagonalization. Furthermore, we develop a time complexity (processing latency) analysis framework with hardware profiling, allowing scalable and realistic evaluation without full implementation. The numerical results demonstrate the superior time efficiency of the selected parallel method, particularly in massive MIMO scenarios.

eess.SP↗

arXiv2025

Serendipitous Recommendation with Multimodal LLM

Conventional recommendation systems succeed in identifying relevant content but often fail to provide users with surprising or novel items. Multimodal Large Language Models (MLLMs) possess the world knowledge and multimodal understanding needed for serendipity, but their integration into billion-item-scale platforms presents significant challenges. In this paper, we propose a novel hierarchical framework where fine-tuned MLLMs provide high-level guidance to conventional recommendation models, steering them towards more serendipitous suggestions. This approach leverages MLLM strengths in understanding multimodal content and user interests while retaining the efficiency of traditional models for item-level recommendation. This mitigates the complexity of applying MLLMs directly to vast action spaces. We also demonstrate a chain-of-thought strategy enabling MLLMs to discover novel user interests by first understanding video content and then identifying relevant yet unexplored interest clusters. Through live experiments within a commercial short-form video platform serving billions of users, we show that our MLLM-powered approach significantly improves both recommendation serendipity and user satisfaction.

cs.IR↗

arXiv2025

Reconfigurable Intelligent Surface-Assisted Multiuser Tracking and Signal Detection in ISAC

This paper investigates the multiuser tracking and signal detection problem in integrated sensing and communication (ISAC) systems with the assistance of reconfigurable intelligent surfaces (RISs). Due to the diverse and high user mobility, the tracking and signal detection performance can be significantly deteriorated without choreographed user state (position and velocity) updating principle. To tackle this challenge, we manage to establish a comprehensive probabilistic signal model to characterize the interdependencies among user states, transmit signals, and received signals during the tracking procedure. Based on the Bayesian problem formulation, we further propose a novel hybrid variational message passing algorithm for the online estimation of user states, which can iteratively update the posterior probabilities of user states during each tracking frame with computational efficiency. Numerical results are provided to demonstrate that the proposed algorithm can significantly improve both of the tracking and signal detection performance over the representative Bayesian estimation counterparts.

eess.SP↗

arXiv2025

Uncertainty-Driven Hierarchical Sampling for Unbalanced Continual Malware Detection with Time-Series Update-Based Retrieval

Android malware detection continues to face persistent challenges stemming from long-term concept drift and class imbalance, as evolving malicious behaviors and shifting usage patterns dynamically reshape feature distributions. Although continual learning (CL) mitigates drift, existing replay-based methods suffer from inherent bias. Specifically, their reliance on classifier uncertainty for sample selection disproportionately prioritizes the dominant benign class, causing overfitting and reduced generalization to evolving malware. To address these limitations, we propose a novel uncertainty-guided CL framework. First, we introduce a hierarchical balanced sampler that employs a dual-phase uncertainty strategy to dynamically balance benign and malicious samples while simultaneously selecting high-information, high-uncertainty instances within each class. This mechanism ensures class equilibrium across both replay and incremental data, thereby enhancing adaptability to emerging threats. Second, we augment the framework with a vector retrieval mechanism that exploits historical malware embeddings to identify evolved variants via similarity-based retrieval, thereby complementing classifier updates. Extensive experiments demonstrate that our framework significantly outperforms state-of-the-art methods under strict low-label conditions (50 labels per phase). It achieves a true positive rate (TPR) of 92.95\% and a mean accuracy (mACC) of 94.26\%, which validates its efficacy for sustainable Android malware detection.

cs.CE↗

arXiv2025

Fabrication and Characterization of the Moiré surface state on a topological insulator

A Moire superlattice on the topological insulator surface is predicted to exhibit many novel properties but has not been experimentally realized. Here, we developed a two-step growth method to successfully fabricate a topological insulator Sb2Te3 thin film with a Moire superlattice, which is generated by a twist of the topmost layer via molecular beam epitaxy. The established Moire topological surface state is characterized by scanning tunneling microscopy and spectroscopy. By application of a magnetic field, new features in Landau levels arise on the Moire region compared to the pristine surface of Sb2Te3, which makes the system a promising platform for pursuing next-generation electronics. Notably, the growth method, which circumvents contamination and the induced interface defects in the manual fabrication method, can be widely applied to other van der Waals materials for fabricating Moire superlattices.

cond-mat.mtrl-sci↗

arXiv2025

Revised classification of the CHIME fast radio bursts with machine learning

Fast radio bursts (FRBs) are short-duration and energetic radio transients of unknown origin. Observationally, they are commonly categorized into repeaters and non-repeaters. However, this binary classification may be influenced by observational limitations such as sensitivity and time coverage of telescopes. In this work, we employ unsupervised machine learning techniques to re-examine the CHIME/FRB catalog, with the goal of identifying intrinsic groupings in the FRB population without relying on preassigned labels. Using t-distributed stochastic neighbor embedding (t-SNE) for dimensionality reduction and hierarchical density-based spatial clustering of applications with noise (HDBSCAN) for clustering, we find that the FRB sample separates naturally into two major clusters. One cluster contains nearly all known repeaters but is contaminated by some apparently non-repeaters, while the other cluster is dominated by non-repeaters. This suggests that certain FRBs previously labeled as non-repeaters may share intrinsic similarities with repeaters. The mutual information analysis reveals that rest-frame frequency width and peak frequency are the most informative features governing the clustering structure. Even when reducing the input space to just these two features, the classification remains robust.

astro-ph.IM↗