arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Author

Ge Sun

48 papers · 2002–2026

2002: 12003: 02004: 02005: 02006: 02007: 02008: 22009: 02010: 02011: 02012: 12013: 42014: 32015: 02016: 22017: 12018: 02019: 02020: 02021: 02022: 02023: 42024: 82025: 122026: 10
2002peak 12 in 20252026

Frequent co-authors

Papers

showing 1–25 of 48 RSS

On generalized multi‐state start‐up demonstration tests

Xian Zhao, Ge Sun, Weijuan Xie +1 more

Statistics and Probability · Applied Stochastic Models in Business and Industry · · cited 17× · DOI ↗ · more like this ·

In view of the fact that start‐up results of products are usually multi‐state, based on the traditional two‐state start‐up demonstration tests, such as consecutive successes total failures (CSTF), total successes consecutive failures (TSCF), consecutive successes consecutive failures (CSCF) and total successes total failures (TSTF), four multi‐state start‐up demonstration tests are proposed in this paper. By using finite Markov chain imbedding approach, the acceptance and rejection probabilities, the probability mass function, the distribution function and the expectation of the number of the start‐up tests to be terminated are given. We also provide a procedure to select the optimal parameter values. Besides, the estimations of possibly unknown probabilities are given by using maximum likelihood estimation. Finally, a numerical example that contains two tables is given to illustrate…

Potential for improved groundwater recharge and dry-season flows through forest landscape restoration on degraded lands in the tropics

L. Adrian Bruijnzeel, Jorge L. Peña-Arancibia, Douglas Sheil +7 more

Artificial Intelligence · Forest Ecosystems · · cited 13× · DOI ↗ · more like this ·

As interest in tropical forest restoration accelerates, understanding its hydrological implications is increasingly urgent. While concerns persist that reforestation will reduce annual water yields—particularly in drier climates—we highlight conditions under which forest landscape restoration (FLR) can improve seasonal water availability, especially during the dry season. We examine the trade-off between increased vegetation water use (“pumping”) and enhanced infiltration and subsurface retention (“sponging”) following forestation of degraded lands, the recovery of vegetation's ability to capture “occult” precipitation (fog) in specific coastal and montane settings, and the role of forest cover in enhancing moisture recycling and transport at multiple scales. A pan-tropical sensitivity analysis shows that in degraded landscapes with deep soils and pronounced rainfall seasonality…

East Asian expatriate football players and national team success: Chinese, Japanese, and South Korean players in Europe (2000–2024)

Luo Le, Yangke Tang, Xiaohan Li +3 more

Gender Studies · Scientific Reports · · cited 2× · DOI ↗ · more like this ·

This study investigated the influence of Chinese, Japanese, and South Korean football players' participation in European leagues on their national teams' FIFA rankings from 2000 to 2024. Utilizing data from 22,972 matches featuring 392 players across 36 European leagues and 12 tournaments or cup competitions, survival and conditional process analyses were conducted to explore the relationships between expatriate player counts, appearances, playing time, and FIFA rankings. The results demonstrated a significant correlation between the number of expatriate players, particularly in top-tier leagues, and national team rankings. Notably, Japanese and South Korean players exhibited longer durations of participation and higher rates of advancement to elite European leagues compared to Chinese players. Furthermore, the conditional process analysis revealed an indirect effect of expatriate…

Sample-Efficient Human Evaluation of Large Language Models via Maximum Discrepancy Competition

Kehua Feng, Keyan Ding, Hongzhi Tan +8 more

cs.LG · · cited 2× · arXiv ↗ · more like this ·

Reliable evaluation of large language models (LLMs) is impeded by two key challenges: objective metrics often fail to reflect human perception of natural language, and exhaustive human labeling is prohibitively expensive. Here, we propose a sample-efficient human evaluation method for LLMs based on the principle of MAximum Discrepancy (MAD) Competition. Our method automatically and adaptively selects a compact set of input instructions that maximize semantic discrepancy between pairs of LLM responses. Human evaluators then perform three-alternative forced choices on these paired responses, which are aggregated into a global ranking using Elo rating. We apply our approach to compare eight widely used LLMs across four tasks: scientific knowledge understanding, mathematical reasoning, creative and functional writing, and code generation and explanation. Experimental results show that our…

DecAP: Decaying Action Priors for Accelerated Imitation Learning of Torque-Based Legged Locomotion Policies

Shivam Sood, Ge Sun, Peizhuo Li +1 more

cs.RO · · cited 2× · arXiv ↗ · more like this ·

Optimal Control for legged robots has gone through a paradigm shift from position-based to torque-based control, owing to the latter's compliant and robust nature. In parallel to this shift, the community has also turned to Deep Reinforcement Learning (DRL) as a promising approach to directly learn locomotion policies for complex real-life tasks. However, most end-to-end DRL approaches still operate in position space, mainly because learning in torque space is often sample-inefficient and does not consistently converge to natural gaits. To address these challenges, we propose a two-stage framework. In the first stage, we generate our own imitation data by training a position-based policy, eliminating the need for expert knowledge to design optimal controllers. The second stage incorporates decaying action priors, a novel method to enhance the exploration of torque-based policies aided…

Introduction: the “vectors” of Chinese history

Viren Murthy, Ge Sun

Sociology and Political Science · Inter-Asia Cultural Studies · · cited 2× · DOI ↗ · more like this ·

This issue contains essays by the important Japanese sinologist and intellectual historian, Mizoguchi Yūzō (1932–2010). Born in Nagoya, he studied in the literature department in Tokyo University and then went to graduate school in Nagoya University. He was a student of Iriya Yoshitaka, the famous sinologist. Mizoguchi has taught at Saitama University, Hitotsubashi University and Tokyo University. He was also a Distinguished Professor at Tokyo University. Mizoguchi Yūzō left a number of important works throughout his life. The selection is unable to provide a complete picture of Mizoguchi’s ideas about the principle of Chinese history, but it provides an entry point for English readers to understand Mizoguchi’s work. In this introduction, we attempt to bring out certain key ideas of Mizoguchi’s writings and discuss his historical and theoretical significance.

The Median Filtering of Digital Image Based on MATLAB

Hai‐Qiang Wang, Xian Ge Sun

Computer Vision and Pattern Recognition · Applied Mechanics and Materials · · cited 2× · DOI ↗ · more like this ·

In the processing of image de-noising, median filtering is a more common nonlinear filtering technique. In this paper, we used the image processing toolbox in the matlab and the adaptive filter program by ourselves. The noise in different intensity of image is reduced. This paper discussed the problem that how we choice of the filtering method under different noise intensity. The results show that median filter can realize effective filter under low noise intensity, if the noise intensity is excessively large, the adaptive filter having a more ideal filtering result than that of standard median filter.

Impacts of Global Change on Water Resources in Dryland East Asia

Ge Sun, Xiaoming Feng, Jingfeng Xiao +9 more

Sociology and Political Science · University of New Hampshire Scholars Repository (University of New Hampshire at Manchester) · · cited 2× · OpenAlex ↗ · more like this ·

The vast Dryland East Asia (DEA) area consists of several large geographic regions including the Qinghai-Tibet Plateau, Loess Plateau, and Mongolia Plateau. T he region is of great importance to the functioning of the earth system under a changing climate. In the past three decades, due to the unprecedented land use/land cover change, urbanization, industrialization and climate change, water stress in many areas in DEA have reached a dangerous level that threatened the sustainability of the region. In addition to reviewing literature for the causes of the water crisis observed in the region, as case studies, we examined water balances at a basin and regional scale using multiple modeling techniques, including a remote sensing-based EC-MOD model, a watershed water balance model (WBMPlus), and an evapotranspiration model calibrated for the Loess Plateau region.

HiPoly: a hierarchical polymer-native AI framework for property prediction and generative design

Ge Sun, Gervasio Zaldivar, Yuan Tian +5 more

physics.chem-ph · · cited 1× · arXiv ↗ · more like this ·

Polymeric materials are central to modern technologies, with applications ranging from energy to health and transportation. Although AI has made significant advances in materials discovery, the hierarchical structure of polymers across multiple length scales makes them inherently difficult to represent in a unified and physically meaningful way. Here we introduce HiPoly, a polymer-native AI framework that processes complete polymer descriptions through a three-level hierarchical graph architecture built on the G2RINS representation. HiPoly encodes stochastic inter-monomer connectivity, composition, and molecular weight directly within its architecture, using physically motivated design principles that mirror the multi-scale nature of polymeric systems. The framework establishes an end-to-end AI-driven workflow from experimental formulation data to property prediction, generative…

PRIMAL3: Pathfinding via Reinforcement and Imitation Multi-Agent Learning - Leveraging LaCAM3

Chengyang He, Tanishq Duhan, Gadiel Sznaier Camps +6 more

cs.RO · · cited 1× · arXiv ↗ · more like this ·

We present PRIMAL3, an ultra-large-scale learning-based framework for multi-agent pathfinding (MAPF) that integrates reinforcement learning, topology-aware communication, LaCAM3-guided training, and PIBT-based action refinement. PRIMAL3 targets failures at topologically critical states, where agents must coordinate decisively around bottlenecks, dead ends, and persistent conflicts. Each agent is represented using features derived from cut vertices, dead-end regions, shortest-path distances, and blocking estimates. Two complementary graphs capture agent interactions: a same-direction following graph propagates multihop context along compatible paths, while a different-direction conflict graph differentiates agents competing for shared space through masked attention and relative features. During training, we propose to let policy entropy identify uncertain agents, for which LaCAM3…

The Athenian Academy: A Seven-Layer Architecture Model for Multi-Agent Systems

Lidong Zhai, Zhijie Qiu, Lvyang Zhang +5 more

cs.MA · · cited 1× · arXiv ↗ · more like this ·

This paper proposes the "Academy of Athens" multi-agent seven-layer framework, aimed at systematically addressing challenges in multi-agent systems (MAS) within artificial intelligence (AI) art creation, such as collaboration efficiency, role allocation, environmental adaptation, and task parallelism. The framework divides MAS into seven layers: multi-agent collaboration, single-agent multi-role playing, single-agent multi-scene traversal, single-agent multi-capability incarnation, different single agents using the same large model to achieve the same target agent, single-agent using different large models to achieve the same target agent, and multi-agent synthesis of the same target agent. Through experimental validation in art creation, the framework demonstrates its unique advantages in task collaboration, cross-scene adaptation, and model fusion. This paper further discusses current…

Comparing and Analysis on Adaptive Information-brokerage Schemes in Wireless Sensor Networks

Xingpo Ma, Guo Song, Lei Li +2 more

Computer Networks and Communications · International Journal of Grid and Distributed Computing · · cited 1× · DOI ↗ · more like this ·

In wireless sensor networks, how producers route and store their own data and how consumers retrieve the desired data efficiently are collectively known as the informationbrokerage problem, which is challenging because of the short communication radius and the limited energy of the sensor nodes. It has been shown in existing works that selecting some sensor nodes as the rendezvous nodes to both store the data and reply to the queries from the consumers is a feasible way to solve such a problem. However, when selecting the optimal rendezvous nodes, most of the existing schemes do not consider the following two factors: the event data generating rates of the producers and the query frequencies of the consumers. In fact, those two factors are very important when determining the optimal locations of the rendezvous nodes because of the dynamic characteristic of the sensor networks. As far as…

Digital Image Hiding Using Iterative Blending in Discrete Cosine Transform Domain

Xin Wang, Xian Ge Sun

Computer Graphics and Computer-Aided Design · Applied Mechanics and Materials · · cited 1× · DOI ↗ · more like this ·

In this paper,the technology about still image copyright protection are analyzed and discussed.We will do the iterative blending calculation of the image using the discrete cosine transform to achieve the hiding of the digital image. In this paper, we show the main idea of the discrete cos transform and the digital image blending at first, then through using the methods of the iterative blending of single image to realize the propose of image hiding. At the same time ,we get the results analysis through the calculation of the pictures ’s PSNR.

ALIGN: Aligning Condition Estimation and Expression in Advantage-Conditioned Behavior Cloning

Zechen Wang, Zhenqi Qiu, Houlong Xiong +8 more

cs.RO · · cited 0× · arXiv ↗ · more like this ·

Human-collected robot manipulation data often mix successful behavior with retries, pauses, recoveries, and failures. Advantage-Conditioned Behavior Cloning (AC-BC) can exploit such mixed-quality data, but only if the conditioning signal is both accurately constructed from offline trajectories and effectively incorporated into the policy's action-vector-field prediction. Existing pipelines can weaken this signal by thresholding estimator-derived advantage scores into noisy conditions, and by routing conditions through prompts that weakly affect the action vector field. We introduce ALIGN, a condition-aligned AC-BC framework for vision-language-action policies. ALIGN directly predicts categorical pairwise conditions in the policy-facing label space and injects them into a DiT action expert through AdaRMS modulation fused with flow time. We evaluate ALIGN on deliberately mixed real-robot…

MIRRA: Fast, High-Fidelity Molecular Generation with Chemistry-Informed Discrete Flow Matching

Ge Sun, Juan J. de Pablo

cs.LG · · cited 0× · arXiv ↗ · more like this ·

Fast, high-fidelity molecular generation requires accurate chemical choices and efficient sampling. Chemical building blocks reduce graph size, but make each node a choice among tens of thousands of related identities. We introduce MIRRA (Molecular Identity Representation, Refresh and Assembly), a discrete flow-matching generator that learns from chemical knowledge and evaluates identity probabilities only when needed, while keeping every identity in its vocabulary available. Chemical features share information across related blocks, and learned identity corrections distinguish them in molecular context. By deciding which positions will refresh before computing their probabilities, the sampler avoids unused predictions while preserving each sampling transition. Against the strongest baselines in each task, MIRRA reduces chemical and graph distribution errors on ZINC250k by 29-31% and…

REACT: Rolling Denoising and Dual Decoupling for Reactive Robot Control with VLA Models

Houlong Xiong, Zhenqi Qiu, Zechen Wang +8 more

cs.RO · · cited 0× · arXiv ↗ · more like this ·

Flow-based vision-language-action (VLA) models generate action chunks for temporally coherent robot motion, but chunked control creates a fundamental closed-loop trade-off: long chunks provide smooth execution, whereas frequent replanning improves reactivity at the cost of action discontinuities. We introduce REACT, a rolling-denoising framework that makes flow-based VLAs more reactive while preserving long-horizon context. Instead of regenerating entire action chunks from scratch, REACT maintains a persistent action buffer with staggered flow timesteps. At each control step, the full horizon is denoised using the latest observation, the cleanest action block is executed, partially refined future blocks are shifted forward, and fresh noise is appended to the tail. As a result, each executed action block is refined across multiple recent observations before deployment. To support…

More applicable form of Beverloo’s equation: Correction of coefficient k

Ge Sun, Quan Chen, Ran Li +5 more

Applied Mathematics · Chemical Engineering Science · · cited 0× · DOI ↗ · more like this ·

Mass discharge rate, as an important characteristic quantity reflecting the fluidity of particulate materials. And it is crucial to accurately measure and calculate the mass discharge rate. There are two issues with the existing Beverloo’s equation: the inability to accurately fit the experiment points within the wide range outlet, and the lack of physical significance for the coefficient k . In this work, the fitting equations and relevant variation law of mass discharge rate were explored for eccentric silos with different sidewall frictional coefficients. The intrinsic connection between coefficient k and diameter D, frictional coefficient μ was revealed for the first time. And the frictional coefficient μ was successfully introduced into Beverloo’s equation, completing the correction and improvement of the equation. By establishing a quantitative relationship between the outlet…

OmniBench: A Comprehensive Benchmark Integrating Real-World, Time-sensitive, and Multi-Hop Questions with a Multi-Dimensional Hybrid Evaluation Framework

Wenjie Wang, Yufeng Jiang, Ge Sun +7 more

Artificial Intelligence · Proceedings of the AAAI Conference on Artificial Intelligence · · cited 0× · DOI ↗ · more like this ·

Recently, with the increasing capabilities of Large Language Models (LLMs), AI applications have gradually emerged to solve various problems in people's daily lives, so accurately measuring their performance and reliability is paramount. However, existing benchmarks predominantly rely on closed-ended, multiple-choice or short-answer question formats. While useful for assessment, these formats exhibit a significant gap compared to the diverse and open-ended nature of questions posed by real-world users. To bridge this gap, we produce OmniBench, a comprehensive open-domain benchmark. OmniBench is uniquely composed of authentic, user-generated questions harvested from real-world interactions on various websites and applications, covering 16 rigorously defined knowledge domains and 5 crucial user intents derived from a large-scale analysis of the mass corpus. Crucially, we propose three…

MGTraj: Multi-Granularity Goal-Guided Human Trajectory Prediction with Recursive Refinement Network

Ge Sun, Jun Ma

cs.CV · · cited 0× · arXiv ↗ · more like this ·

Accurate human trajectory prediction is crucial for robotics navigation and autonomous driving. Recent research has demonstrated that incorporating goal guidance significantly enhances prediction accuracy by reducing uncertainty and leveraging prior knowledge. Most goal-guided approaches decouple the prediction task into two stages: goal prediction and subsequent trajectory completion based on the predicted goal, which operate at extreme granularities: coarse-grained goal prediction forecasts the overall intention, while fine-grained trajectory completion needs to generate the positions for all future timesteps. The potential utility of intermediate temporal granularity remains largely unexplored, which motivates multi-granularity trajectory modeling. While prior work has shown that multi-granularity representations capture diverse scales of human dynamics and motion patterns…

DragTraffic: Interactive and Controllable Traffic Scene Generation for Autonomous Driving

Sheng Wang, Ge Sun, Fulong Ma +5 more

cs.RO · · cited 0× · arXiv ↗ · more like this ·

Evaluating and training autonomous driving systems require diverse and scalable corner cases. However, most existing scene generation methods lack controllability, accuracy, and versatility, resulting in unsatisfactory generation results. Inspired by DragGAN in image generation, we propose DragTraffic, a generalized, interactive, and controllable traffic scene generation framework based on conditional diffusion. DragTraffic enables non-experts to generate a variety of realistic driving scenarios for different types of traffic agents through an adaptive mixture expert architecture. We employ a regression model to provide a general initial solution and a refinement process based on the conditional diffusion model to ensure diversity. User-customized context is introduced through cross-attention to ensure high controllability. Experiments on a real-world driving dataset show that…

Multi-feature Interaction for Aspect Sentiment Triplet Extraction

CHEN Linying, LIU Jianhua, ZHENG Zhixiong, LIN Jie, XU Ge, SUN Shuihua

Artificial Intelligence · DOAJ (DOAJ: Directory of Open Access Journals) · · cited 0× · DOI ↗ · more like this ·

Aspect sentiment triple extraction is one of the subtasks of aspect-level sentiment analysis, which aims to extract aspect terms, corresponding opinion terms and sentiment polarity in sentence. Previous studies focus on designing a new paradigm to complete the triplet extraction task in an end-to-end manner. They ignore the role of external knowledge in the model, thus semantic information, part-of-speech information and local context information are not fully explored and utilized. Aiming at the above problems, multi-feature interaction for aspect sentiment triplet extraction (MFI-ASTE) is proposed. Firstly, the bidirectional encoder representation from transformers (BERT) model is used to learn the context semantic feature information, meanwhile, the self-attention mechanism is used to strengthen the semantic feature. Secondly, the semantic feature interacts with the extracted…

Facing history with a sense of awe: an interview with Professor Mizoguchi Yūzō

Kuan‐Hsing Chen, Ge Sun

Cultural Studies · Inter-Asia Cultural Studies · · cited 0× · DOI ↗ · more like this ·

In this wide-ranging interview and discussion with Kuan-Hsing Chen and Sun Ge, Mizoguchi Yūzō describes the origins of his interest in China studies and the process through which he developed his perspective on China, Japan, and the world. Mizoguchi details his break with both old-style Japanese Sinology and Western-influenced scholarship, which assumes Japanese superiority over China and takes Euro-American society and concepts as its standard. Mizoguchi suggests that historians can and should cultivate a new subjectivity for themselves and understanding of the history of the world as a whole through an approach to China that attempts to understand China’s own internal historical processes rather than assuming the universality of Western processes. He discusses his efforts to help reform the institutional structure of China studies in Japan, and further touches on the part played by…

(wileyonlinelibrary.com) DOI: 10.1002/eco.200 Effects of Spring Drought on Carbon Sequestration, Evapotranspiration and Water Use Efficiency in the Songnen Meadow Steppe in Northeast

Gang Dong, Jixun Guo, Jiquan Chen +4 more

Sociology and Political Science · · cited 0× · OpenAlex ↗ · more like this ·

Global climate change projections suggest an increasing frequency of droughts and extreme rain events in the steppes of the Eurasian region. Using the eddy covariance method, we measured carbon and water balances of a meadow steppe ecosystem in Northeast China during 2 years which had contrasting precipitation patterns in spring seasons in 2007 and 2008. The meadow steppe sequestrated only 64Ð2 gCm 2 year 1 in 2007 compared to 160Ð5 gCm 2 year 1 in 2008, due to a severe spring drought in 2007. The 2007 spring drought resulted in a dramatic reduction of leaf area index (LAI) and aboveground net primary productivity (ANPP). However, the meadow steppe still acted as a carbon sink in 2007. The strength of the sink was

A COMPARISON OF MIKE SHE AND DRAINMOD FOR MODELING FORESTED WETLAND HYDROLOGY IN COASTAL SOUTH CAROLINA,

Zhaohua Dai, Devendra M. Amatya, Ge Sun +3 more

Political Science and International Relations · · cited 0× · OpenAlex ↗ · more like this ·

ABSTRACT Models are widely used to assess hydrologic impacts of land-management, land-use change and climate change. Two hydrologic models with different spatial scales, MIKE SHE (spatially distributed, watershed-scale) and DRAINMOD (lumped, fieldscale), were compared in terms of their performance in predicting stream flow and water table depth in a first-order forested watershed in coastal South Carolina. The model performance was evaluated using the coefficient of determination (R 2) and Nash-Sutcliffe’s model efficiency (E). Although both models performed reasonably well in predicting monthly and annual average water table depths and stream flow with acceptable E values (0.55-0.99) for the five-year period (2003-2007), MIKE SHE yielded better results than DRAINMOD for daily hydrologic dynamics. Both models, however, showed relatively large uncertainty in simulating stream flow for…