arXiv subjects

Wei Gao

Publications and source records attributed to Wei Gao.

At least 145 records · Page 8Linked to original sources

arXiv2025

High-Precision Transformer-Based Visual Servoing for Humanoid Robots in Aligning Tiny Objects

High-precision tiny object alignment remains a common and critical challenge for humanoid robots in real-world. To address this problem, this paper proposes a vision-based framework for precisely estimating and controlling the relative position between a handheld tool and a target object for humanoid robots, e.g., a screwdriver tip and a screw head slot. By fusing images from the head and torso cameras on a robot with its head joint angles, the proposed Transformer-based visual servoing method can correct the handheld tool's positional errors effectively, especially at a close distance. Experiments on M4-M8 screws demonstrate an average convergence error of 0.8-1.3 mm and a success rate of 93\%-100\%. Through comparative analysis, the results validate that this capability of high-precision tiny object alignment is enabled by the Distance Estimation Transformer architecture and the Multi-Perception-Head mechanism proposed in this paper.

cs.CV↗

arXiv2025

Data Can Speak for Itself: Quality-guided Utilization of Wireless Synthetic Data

Generative models have gained significant attention for their ability to produce realistic synthetic data that supplements the quantity of real-world datasets. While recent studies show performance improvements in wireless sensing tasks by incorporating all synthetic data into training sets, the quality of synthetic data remains unpredictable and the resulting performance gains are not guaranteed. To address this gap, we propose tractable and generalizable metrics to quantify quality attributes of synthetic data - affinity and diversity. Our assessment reveals prevalent affinity limitation in current wireless synthetic data, leading to mislabeled data and degraded task performance. We attribute the quality limitation to generative models' lack of awareness of untrained conditions and domain-specific processing. To mitigate these issues, we introduce SynCheck, a quality-guided synthetic data utilization scheme that refines synthetic data quality during task model training. Our evaluation demonstrates that SynCheck consistently outperforms quality-oblivious utilization of synthetic data, and achieves 4.3% performance improvement even when the previous utilization degrades performance by 13.4%.

cs.LG↗

arXiv2025

Stochasticity-aware No-Reference Point Cloud Quality Assessment

The evolution of point cloud processing algorithms necessitates an accurate assessment for their quality. Previous works consistently regard point cloud quality assessment (PCQA) as a MOS regression problem and devise a deterministic mapping, ignoring the stochasticity in generating MOS from subjective tests. This work presents the first probabilistic architecture for no-reference PCQA, motivated by the labeling process of existing datasets. The proposed method can model the quality judging stochasticity of subjects through a tailored conditional variational autoencoder (CVAE) and produces multiple intermediate quality ratings. These intermediate ratings simulate the judgments from different subjects and are then integrated into an accurate quality prediction, mimicking the generation process of a ground truth MOS. Specifically, our method incorporates a Prior Module, a Posterior Module, and a Quality Rating Generator, where the former two modules are introduced to model the judging stochasticity in subjective tests, while the latter is developed to generate diverse quality ratings. Extensive experiments indicate that our approach outperforms previous cutting-edge methods by a large margin and exhibits gratifying cross-dataset robustness. Codes are available at https://git.openi.org.cn/OpenPointCloud/nrpcqa.

cs.CV↗

arXiv2025

Study of Stability and Consistency of EAS Thermal Neutron Detection at ENDA-64

Introduction:Electron-Neutron Detector Array (ENDA) is designed to measure thermal neutrons produced by hadronic interactions between cosmic ray extensive air showers (EAS) and the surrounding environment as well as electrons around the cores of EAS. ENDA is located within Large High Altitude Air Shower Observatory (LHAASO). ENDA was expanded from an initial 16 detectors to 64 detectors in April 2023, so called ENDA-64, and has been running alongside LHAASO. The stability and consistency of neutron detection are crucial for laying a solid foundation for subsequent data analysis and physical results. Methods:We obtain the stability by studying variations of event rate and thermal neutron rate in each cluster and the consistency by comparing distribution of number of thermal neutrons between clusters. Additionally, we investigate the specific influences of the rainy and dry seasons, as well as the presence or absence of sand cubes under the detectors, to examine the environmental factors affecting neutron measurement performance. Results:The calibration results indicate good consistency in thermal neutron detection across the clusters, with the maximum inconsistency of 6.85%. The maximum instability of event rate and thermal neutron rate over time are 4.68% and 11.0% respectively. The maximum inconsistency between the clusters without the sand cubes is 18%. The use of sand cubes is effective in protecting the target material from rainwater, and the sand cubes help the cluster to increase collection of neutrons generated by EAS events.

hep-ex↗

arXiv2025

Generalized Gaussian Entropy Model for Point Cloud Attribute Compression with Dynamic Likelihood Intervals

Gaussian and Laplacian entropy models are proved effective in learned point cloud attribute compression, as they assist in arithmetic coding of latents. However, we demonstrate through experiments that there is still unutilized information in entropy parameters estimated by neural networks in current methods, which can be used for more accurate probability estimation. Thus we introduce generalized Gaussian entropy model, which controls the tail shape through shape parameter to more accurately estimate the probability of latents. Meanwhile, to the best of our knowledge, existing methods use fixed likelihood intervals for each integer during arithmetic coding, which limits model performance. We propose Mean Error Discriminator (MED) to determine whether the entropy parameter estimation is accurate and then dynamically adjust likelihood intervals. Experiments show that our method significantly improves rate-distortion (RD) performance on three VAE-based models for point cloud attribute compression, and our method can be applied to other compression tasks, such as image and video compression.

cs.CV↗

arXiv2025

Universal machine learning interatomic potentials poised to supplant DFT in modeling general defects in metals and random alloys

Recent advances in machine learning, combined with the generation of extensive density functional theory (DFT) datasets, have enabled the development of universal machine learning interatomic potentials (uMLIPs). These models offer broad applicability across the periodic table, achieving first-principles accuracy at a fraction of the computational cost of traditional DFT calculations. In this study, we demonstrate that state-of-the-art pretrained uMLIPs can effectively replace DFT for accurately modeling complex defects in a wide range of metals and alloys. Our investigation spans diverse scenarios, including grain boundaries and general defects in pure metals, defects in high-entropy alloys, hydrogen-alloy interactions, and solute-defect interactions. Remarkably, the latest EquiformerV2 models achieve DFT-level accuracy on comprehensive defect datasets, with root mean square errors (RMSE) below 5 meV/atom for energies and 100 meV/Å for forces, outperforming specialized machine learning potentials such as moment tensor potential and atomic cluster expansion. We also present a systematic analysis of accuracy versus computational cost and explore uncertainty quantification for uMLIPs. A detailed case study of tungsten (W) demonstrates that data on pure W alone is insufficient for modeling complex defects in uMLIPs, underscoring the critical importance of advanced machine learning architectures and diverse datasets, which include over 100 million structures spanning all elements. These findings establish uMLIPs as a robust alternative to DFT and a transformative tool for accelerating the discovery and design of high-performance materials.

cond-mat.mtrl-sci↗

arXiv2025

Reinforcement Learning Optimization for Large-Scale Learning: An Efficient and User-Friendly Scaling Library

We introduce ROLL, an efficient, scalable, and user-friendly library designed for Reinforcement Learning Optimization for Large-scale Learning. ROLL caters to three primary user groups: tech pioneers aiming for cost-effective, fault-tolerant large-scale training, developers requiring flexible control over training workflows, and researchers seeking agile experimentation. ROLL is built upon several key modules to serve these user groups effectively. First, a single-controller architecture combined with an abstraction of the parallel worker simplifies the development of the training pipeline. Second, the parallel strategy and data transfer modules enable efficient and scalable training. Third, the rollout scheduler offers fine-grained management of each sample's lifecycle during the rollout stage. Fourth, the environment worker and reward worker support rapid and flexible experimentation with agentic RL algorithms and reward designs. Finally, AutoDeviceMapping allows users to assign resources to different models flexibly across various stages.

cs.LG↗

arXiv2025

CausalAbstain: Enhancing Multilingual LLMs with Causal Reasoning for Trustworthy Abstention

Large Language Models (LLMs) often exhibit knowledge disparities across languages. Encouraging LLMs to \textit{abstain} when faced with knowledge gaps is a promising strategy to reduce hallucinations in multilingual settings. Current abstention strategies for multilingual scenarios primarily rely on generating feedback in various languages using LLMs and performing self-reflection. However, these methods can be adversely impacted by inaccuracies and biases in the generated feedback. To address this, from a causal perspective, we introduce \textit{CausalAbstain}, a method that helps LLMs determine whether to utilize multiple generated feedback responses and how to identify the most useful ones. Extensive experiments demonstrate that \textit{CausalAbstain} effectively selects helpful feedback and enhances abstention decisions with interpretability in both native language (\textsc{Casual-native}) and multilingual (\textsc{Causal-multi}) settings, outperforming strong baselines on two benchmark datasets covering encyclopedic and commonsense knowledge QA tasks. Our code and data are open-sourced at https://github.com/peachch/CausalAbstain.

cs.CL↗

arXiv2025

A Novel Benchmark and Dataset for Efficient 3D Gaussian Splatting with Gaussian Point Cloud Compression

Recently, immersive media and autonomous driving applications have significantly advanced through 3D Gaussian Splatting (3DGS), which offers high-fidelity rendering and computational efficiency. Despite these advantages, 3DGS as a display-oriented representation requires substantial storage due to its numerous Gaussian attributes. Current compression methods have shown promising results but typically neglect the compression of Gaussian spatial positions, creating unnecessary bitstream overhead. We conceptualize Gaussian primitives as point clouds and propose leveraging point cloud compression techniques for more effective storage. AI-based point cloud compression demonstrates superior performance and faster inference compared to MPEG Geometry-based Point Cloud Compression (G-PCC). However, direct application of existing models to Gaussian compression may yield suboptimal results, as Gaussian point clouds tend to exhibit globally sparse yet locally dense geometric distributions that differ from conventional point cloud characteristics. To address these challenges, we introduce GausPcgc for Gaussian point cloud geometry compression along with a specialized training dataset GausPcc-1K. Our work pioneers the integration of AI-based point cloud compression into Gaussian compression pipelines, achieving superior compression ratios. The framework complements existing Gaussian compression methods while delivering significant performance improvements. All code, data, and pre-trained models will be publicly released to facilitate further research advances in this field.

cs.GR↗

arXiv2025

ProMind-LLM: Proactive Mental Health Care via Causal Reasoning with Sensor Data

Mental health risk is a critical global public health challenge, necessitating innovative and reliable assessment methods. With the development of large language models (LLMs), they stand out to be a promising tool for explainable mental health care applications. Nevertheless, existing approaches predominantly rely on subjective textual mental records, which can be distorted by inherent mental uncertainties, leading to inconsistent and unreliable predictions. To address these limitations, this paper introduces ProMind-LLM. We investigate an innovative approach integrating objective behavior data as complementary information alongside subjective mental records for robust mental health risk assessment. Specifically, ProMind-LLM incorporates a comprehensive pipeline that includes domain-specific pretraining to tailor the LLM for mental health contexts, a self-refine mechanism to optimize the processing of numerical behavioral data, and causal chain-of-thought reasoning to enhance the reliability and interpretability of its predictions. Evaluations of two real-world datasets, PMData and Globem, demonstrate the effectiveness of our proposed methods, achieving substantial improvements over general LLMs. We anticipate that ProMind-LLM will pave the way for more dependable, interpretable, and scalable mental health case solutions.

cs.AI↗

arXiv2025

Never Start from Scratch: Expediting On-Device LLM Personalization via Explainable Model Selection

Personalization of Large Language Models (LLMs) is important in practical applications to accommodate the individual needs of different mobile users. Due to data privacy concerns, LLM personalization often needs to be locally done at the user's mobile device, but such on-device personalization is constrained by both the limitation of on-device compute power and insufficiency of user's personal data. In this paper, we address these constraints by fine-tuning an already personalized LLM with user's personal data, and present XPerT, a new technique that ensure proper selection of such already personalized LLMs based on explainability about how they were being fine-tuned. We implemented and evaluated XPerT on various smartphone models with mainstream LLMs, and experiment results show that XPerT reduces the computation costs of on-device LLM personalization by 83%, and improves its data efficiency by 51%.

cs.LG↗

arXiv2025

ClarityEthic: Explainable Moral Judgment Utilizing Contrastive Ethical Insights from Large Language Models

With the rise and widespread use of Large Language Models (LLMs), ensuring their safety is crucial to prevent harm to humans and promote ethical behaviors. However, directly assessing value valence (i.e., support or oppose) by leveraging large-scale data training is untrustworthy and inexplainable. We assume that emulating humans to rely on social norms to make moral decisions can help LLMs understand and predict moral judgment. However, capturing human values remains a challenge, as multiple related norms might conflict in specific contexts. Consider norms that are upheld by the majority and promote the well-being of society are more likely to be accepted and widely adopted (e.g., "don't cheat,"). Therefore, it is essential for LLM to identify the appropriate norms for a given scenario before making moral decisions. To this end, we introduce a novel moral judgment approach called \textit{ClarityEthic} that leverages LLMs' reasoning ability and contrastive learning to uncover relevant social norms for human actions from different perspectives and select the most reliable one to enhance judgment accuracy. Extensive experiments demonstrate that our method outperforms state-of-the-art approaches in moral judgment tasks. Moreover, human evaluations confirm that the generated social norms provide plausible explanations that support the judgments. This suggests that modeling human moral judgment with the emulating humans moral strategy is promising for improving the ethical behaviors of LLMs.

cs.CY↗

arXiv2025

PhyT2V: LLM-Guided Iterative Self-Refinement for Physics-Grounded Text-to-Video Generation

Text-to-video (T2V) generation has been recently enabled by transformer-based diffusion models, but current T2V models lack capabilities in adhering to the real-world common knowledge and physical rules, due to their limited understanding of physical realism and deficiency in temporal modeling. Existing solutions are either data-driven or require extra model inputs, but cannot be generalizable to out-of-distribution domains. In this paper, we present PhyT2V, a new data-independent T2V technique that expands the current T2V model's capability of video generation to out-of-distribution domains, by enabling chain-of-thought and step-back reasoning in T2V prompting. Our experiments show that PhyT2V improves existing T2V models' adherence to real-world physical rules by 2.3x, and achieves 35% improvement compared to T2V prompt enhancers. The source codes are available at: https://github.com/pittisl/PhyT2V.

cs.CV↗

arXiv2025

Rethinking Key-Value Cache Compression Techniques for Large Language Model Serving

Key-Value cache (\texttt{KV} \texttt{cache}) compression has emerged as a promising technique to optimize Large Language Model (LLM) serving. It primarily decreases the memory consumption of \texttt{KV} \texttt{cache} to reduce the computation cost. Despite the development of many compression algorithms, their applications in production environments are still not prevalent. In this paper, we revisit mainstream \texttt{KV} \texttt{cache} compression solutions from a practical perspective. Our contributions are three-fold. First, we comprehensively review existing algorithmic designs and benchmark studies for \texttt{KV} \texttt{cache} compression and identify missing pieces in their performance measurement, which could hinder their adoption in practice. Second, we empirically evaluate representative \texttt{KV} \texttt{cache} compression methods to uncover two key issues that affect the computational efficiency: (1) while compressing \texttt{KV} \texttt{cache} can reduce memory consumption, current implementations (e.g., FlashAttention, PagedAttention) do not optimize for production-level LLM serving, resulting in suboptimal throughput performance; (2) compressing \texttt{KV} \texttt{cache} may lead to longer outputs, resulting in increased end-to-end latency. We further investigate the accuracy performance of individual samples rather than the overall performance, revealing the intrinsic limitations in \texttt{KV} \texttt{cache} compression when handling specific LLM tasks. Third, we provide tools to shed light on future \texttt{KV} \texttt{cache} compression studies and facilitate their practical deployment in production. They are open-sourced in \href{https://github.com/LLMkvsys/rethink-kv-compression}{https://github.com/LLMkvsys/rethink-kv-compression}.

cs.LG↗

arXiv2025

Learning Orientation Field for OSM-Guided Autonomous Navigation

OpenStreetMap (OSM) has gained popularity recently in autonomous navigation due to its public accessibility, lower maintenance costs, and broader geographical coverage. However, existing methods often struggle with noisy OSM data and incomplete sensor observations, leading to inaccuracies in trajectory planning. These challenges are particularly evident in complex driving scenarios, such as at intersections or facing occlusions. To address these challenges, we propose a robust and explainable two-stage framework to learn an Orientation Field (OrField) for robot navigation by integrating LiDAR scans and OSM routes. In the first stage, we introduce the novel representation, OrField, which can provide orientations for each grid on the map, reasoning jointly from noisy LiDAR scans and OSM routes. To generate a robust OrField, we train a deep neural network by encoding a versatile initial OrField and output an optimized OrField. Based on OrField, we propose two trajectory planners for OSM-guided robot navigation, called Field-RRT* and Field-Bezier, respectively, in the second stage by improving the Rapidly Exploring Random Tree (RRT) algorithm and Bezier curve to estimate the trajectories. Thanks to the robustness of OrField which captures both global and local information, Field-RRT* and Field-Bezier can generate accurate and reliable trajectories even in challenging conditions. We validate our approach through experiments on the SemanticKITTI dataset and our own campus dataset. The results demonstrate the effectiveness of our method, achieving superior performance in complex and noisy conditions. Our code for network training and real-world deployment is available at https://github.com/IMRL/OriField.

cs.RO↗

arXiv2025

UniPCGC: Towards Practical Point Cloud Geometry Compression via an Efficient Unified Approach

Learning-based point cloud compression methods have made significant progress in terms of performance. However, these methods still encounter challenges including high complexity, limited compression modes, and a lack of support for variable rate, which restrict the practical application of these methods. In order to promote the development of practical point cloud compression, we propose an efficient unified point cloud geometry compression framework, dubbed as UniPCGC. It is a lightweight framework that supports lossy compression, lossless compression, variable rate and variable complexity. First, we introduce the Uneven 8-Stage Lossless Coder (UELC) in the lossless mode, which allocates more computational complexity to groups with higher coding difficulty, and merges groups with lower coding difficulty. Second, Variable Rate and Complexity Module (VRCM) is achieved in the lossy mode through joint adoption of a rate modulation module and dynamic sparse convolution. Finally, through the dynamic combination of UELC and VRCM, we achieve lossy compression, lossless compression, variable rate and complexity within a unified framework. Compared to the previous state-of-the-art method, our method achieves a compression ratio (CR) gain of 8.1\% on lossless compression, and a Bjontegaard Delta Rate (BD-Rate) gain of 14.02\% on lossy compression, while also supporting variable rate and variable complexity.

cs.CV↗

arXiv2025

SPPO:Efficient Long-sequence LLM Training via Adaptive Sequence Pipeline Parallel Offloading

In recent years, Large Language Models (LLMs) have exhibited remarkable capabilities, driving advancements in real-world applications. However, training LLMs on increasingly long input sequences imposes significant challenges due to high GPU memory and computational demands. Existing solutions face two key limitations: (1) memory reduction techniques, such as activation recomputation and CPU offloading, compromise training efficiency; (2) distributed parallelism strategies require excessive GPU resources, limiting the scalability of input sequence length. To address these gaps, we propose Adaptive Sequence Pipeline Parallel Offloading (SPPO), a novel LLM training framework that optimizes memory and computational resource efficiency for long-sequence training. SPPO introduces adaptive offloading, leveraging sequence-aware offloading, and two-level activation management to reduce GPU memory consumption without degrading the training efficiency. Additionally, SPPO develops an adaptive pipeline scheduling approach with a heuristic solver and multiplexed sequence partitioning to improve computational resource efficiency. Experimental results demonstrate that SPPO achieves up to 3.38x throughput improvement over Megatron-LM and DeepSpeed, realizing efficient training of a 7B LLM with sequence lengths of up to 4M tokens on only 128 A100 GPUs.

cs.DC↗

arXiv2025

LLM-Enhanced Multiple Instance Learning for Joint Rumor and Stance Detection with Social Context Information

The proliferation of misinformation, such as rumors on social media, has drawn significant attention, prompting various expressions of stance among users. Although rumor detection and stance detection are distinct tasks, they can complement each other. Rumors can be identified by cross-referencing stances in related posts, and stances are influenced by the nature of the rumor. However, existing stance detection methods often require post-level stance annotations, which are costly to obtain. We propose a novel LLM-enhanced MIL approach to jointly predict post stance and claim class labels, supervised solely by claim labels, using an undirected microblog propagation model. Our weakly supervised approach relies only on bag-level labels of claim veracity, aligning with multi-instance learning (MIL) principles. To achieve this, we transform the multi-class problem into multiple MIL-based binary classification problems. We then employ a discriminative attention layer to aggregate the outputs from these classifiers into finer-grained classes. Experiments conducted on three rumor datasets and two stance datasets demonstrate the effectiveness of our approach, highlighting strong connections between rumor veracity and expressed stances in responding posts. Our method shows promising performance in joint rumor and stance detection compared to the state-of-the-art methods.

cs.CL↗