arXiv subjects

Chenxi Sun

Publications and source records attributed to Chenxi Sun.

At least 19 recordsLinked to original sources

arXiv2026

Optimal Pricing and Charging Strategy Design for Non-cooperative Battery Swapping Stations

Battery swapping is a rapid way to recharge electric vehicles (EVs). As more and more entities are involved in building Battery Swapping Stations (BSSs), how non-cooperative BSSs maximize their profit in a competitive market needs further investigation. In this paper, we focus on a practical scenario where competitive BSSs are coordinated by the same aggregator. To study the optimal pricing and battery charging, we formulate a hierarchical game-theoretic model, where BSSs determine the swapping price in the day-ahead market in the first stage, and then determine the optimal battery charging strategy in the real-time market in the second stage. We rigorously prove the existence and uniqueness of the Subgame Perfect Nash Equilibrium (SPNE). In particular, the uniqueness property provides theoretical support that the strategy under equilibrium is optimal in the competitive environment. Based on the unique SPNE, we propose an optimal pricing and charging strategy for each BSS to maximize profit in the competitive market. A prediction error handling method is also proposed to deal with unexpected fluctuations in swapping demand. Our simulation with a 12-BSS system based on real-life data from Xi'an, China shows that our pricing and charging strategy increases the individual BSS profit by at least 18.1\%, while the optimal charging strategy naturally achieves peak shaving for the power grid.

cs.GT↗

arXiv2026

HermesHFL: Incentive-Compatible Hierarchical Federated Unlearning for Dynamic LLM Fine-Tuning

Hierarchical federated unlearning (HFUL) for large language model (LLM) fine-tuning faces significant challenges due to hierarchical aggregation, dynamic client participation, and strong parameter coupling in LLM adaptation. Selectively removing client contributions is particularly difficult because model updates propagate across multiple aggregation stages while unlearning requests may coincide with client departures and rejoining. To address these issues, we propose HermesHFL, a hierarchical federated learning framework that supports selective unlearning, dynamic client participation, and client reintegration for scalable LLM fine-tuning via parameter-efficient fine-tuning (PEFT) with LoRA. We formulate a unified optimization problem that jointly models client participation, edge association, incentive allocation, and unlearning under heterogeneous client behaviors. To solve this problem efficiently, we develop Neogen, a neural-guided bilevel evolutionary optimization framework that combines CMA-ES for continuous incentive optimization with a CHC-based evolutionary mechanism for discrete participation and association decisions. A neural surrogate further accelerates optimization and improves search efficiency. Extensive experiments on LLM fine-tuning tasks demonstrate that HermesHFL consistently outperforms state-of-the-art baselines in model utility, unlearning effectiveness, convergence stability, and resource efficiency.

cs.CE↗

arXiv2026

OSOR: One-Step Diffusion Inpainting for Effect-Aware Object Removal

Real-world object removal is challenging due to two key difficulties: the target object's non-local effects, such as shadows and reflections, which are difficult to model, and the fact that user-provided masks are often inaccurate or incomplete. With billions of parameters and tens of denoising steps, diffusion-based models achieve strong removal performance at the expense of substantial computational cost, limiting their use in interactive applications and on edge devices. To address these challenges, we present OSOR (One-Step Object Removal), which simultaneously achieves efficient, effect-aware, and mask-robust object removal. Concretely, OSOR introduces: (1) an occupancy-guided discriminator for precise boundary supervision, enabling stable single-step diffusion training; (2) an alpha head that leverages knowledge from pretrained diffusion models to predict appropriate removal regions with minimal overhead, thereby handling imperfect masks; and (3) a semantic-anchored verification pipeline (SAVP) that filters noisy instruction-based triplets to produce effect-aware supervision at scale. Using SAVP, we curate CORNE, which contains 280K verified removal pairs, and further annotate AnimeEraseBench and TextEraseBench to evaluate performance on more complex removal tasks. Experiments show that OSOR surpasses strong multi-step diffusion baselines in perceptual quality while achieving $4\times$ to $30\times$ faster inference.

cs.CV↗

arXiv2026

Collective response and noise of a levitated ferromagnet lattice for ultralight dark matter detection

Ultralight dark matter can induce weak oscillating magnetic-like signals and can therefore be searched for with precision magnetometry. Levitated ferromagnets provide a sensitive platform for such searches, but a single ferromagnet is limited in total polarized spin and readout performance. We investigate a levitated ferromagnet lattice as a scalable detector for ultralight dark matter. We develop a theoretical description of the collective lattice response in the fully trapped regime, incorporating dipole-dipole interactions, finite-size effects, and boundary-induced mode mixing. We further analyze the collective noise budget and show that interaction effects mainly produce a narrow blind zone through thermal-noise amplification, while away from this region, the lattice preserves favorable collective noise scaling. We then derive projected sensitivities to axion-electron, dark-photon, and axion-photon couplings. We find that the lattice improves the reach in all three channels relative to a single-ferromagnet detector, with an additional coherent signal enhancement in the axion-photon channel from the lattice-generated electromagnetic background.

hep-ph↗

arXiv2026

Boundary-supported radial layering in Hoag-like ring galaxies

Clean Hoag-like ring galaxies are often characterized by an old compact central component, a depleted gap, and a detached outer ring. We identify a boundary-supported radial-layering mechanism in a shell-deformed Kepler control model. A compact inner boundary supplies the core state, while a localized effective shell deformation, interpreted as the reduced imprint of externally supplied material settled near a finite circularization radius, needs to create only an internal maximum and a subsequent outer minimum. These act as the gap barrier and ring-supporting well. The onset of this structure is organized by a saddle-node threshold of the critical-point equation. In a 10^4-point Monte Carlo scan, shell-localized boundary-supported candidates occupy finite parameter volume under the adopted priors, and none of the localized candidates contains an ordered interior minimum--maximum--minimum subsequence. The same branch gives a scale-free gap-to-ring interval overlapping representative ratios for Hoag's Object, UGC 4599, and PGC 1000714, but not for the environmentally processed comparison object JO171.

astro-ph.GA↗

arXiv2026

mtslearn: Machine Learning in Python for Medical Time Series

Medical time-series data captures the dynamic progression of patient conditions, playing a vital role in modern clinical decision support systems. However, real-world clinical data is highly heterogeneous and inconsistently formatted. Furthermore, existing machine learning tools often have steep learning curves and fragmented workflows. Consequently, a significant gap remains between cutting-edge AI technologies and clinical application. To address this, we introduce mtslearn, an end-to-end integrated toolkit specifically designed for medical time-series data. First, the framework provides a unified data interface that automates the parsing and alignment of wide, long, and flat data formats. This design significantly reduces data cleaning overhead. Building on this, mtslearn provides a complete pipeline from data reading and feature engineering to model training and result visualization. Furthermore, it offers flexible interfaces for custom algorithms. Through a modular design, mtslearn simplifies complex data engineering tasks into a few lines of code. This significantly lowers the barrier to entry for clinicians with limited programming experience, empowering them to focus more on exploring medical hypotheses and accelerating the translation of advanced algorithms into real-world clinical practice. mtslearn is publicly available at https://github.com/PKUDigitalHealth/mtslearn.

cs.LG↗

arXiv2026

WebTestBench: Evaluating Computer-Use Agents towards End-to-End Automated Web Testing

The emergence of Large Language Models (LLMs) has catalyzed a paradigm shift in programming, giving rise to "vibe coding", where users can build complete projects and even control computers using natural language instructions. This paradigm has driven automated webpage development, but it introduces a new requirement about how to automatically verify whether the web functionalities are reliably implemented. Existing works struggle to adapt, relying on static visual similarity or predefined checklists that constrain their utility in open-ended environments. Furthermore, they overlook a vital aspect of software quality, namely latent logical constraints. To address these gaps, we introduce WebTestBench, a benchmark for evaluating end-to-end automated web testing. WebTestBench encompasses comprehensive dimensions across diverse web application categories. We decompose the testing process into two cascaded sub-tasks, checklist generation and defect detection, and propose WebTester, a baseline framework for this task. Evaluating popular LLMs with WebTester reveals severe challenges, including insufficient test completeness, detection bottlenecks, and long-horizon interaction unreliability. These findings expose a substantial gap between current computer-use agent capabilities and industrial-grade deployment demands. We hope that WebTestBench provides valuable insights and guidance for advancing end-to-end automated web testing. Our dataset and code are available at https://github.com/friedrichor/WebTestBench.

cs.SE↗

arXiv2026

Structurally Aligned Subtask-Level Memory for Software Engineering Agents

Large Language Models (LLMs) have demonstrated significant potential as autonomous software engineering (SWE) agents. Recent work has further explored augmenting these agents with memory mechanisms to support long-horizon reasoning. However, these approaches typically operate at a coarse instance granularity, treating the entire problem-solving episode as the atomic unit of storage and retrieval. We empirically demonstrate that instance-level memory suffers from a fundamental granularity mismatch, resulting in misguided retrieval when tasks with similar surface descriptions require distinct reasoning logic at specific stages. To address this, we propose Structurally Aligned Subtask-Level Memory, a method that aligns memory storage, retrieval, and updating with the agent's functional decomposition. Extensive experiments on SWE-bench Verified demonstrate that our method consistently outperforms both vanilla agents and strong instance-level memory baselines across diverse backbones, improving mean Pass@1 over the vanilla agent by +4.7 pp on average (e.g., +6.8 pp on Gemini 2.5 Pro). Performance gains grow with more interaction steps, showing that leveraging past experience benefits long-horizon reasoning in complex software engineering tasks.

cs.SE↗

arXiv2026

Ultralight Dark Matter Detection with a Ferromagnet Lattice

A levitated ferromagnet provides an exceptionally sensitive probe of ultralight dark matter (ULDM) through measuring weak magnetic-like field signals. We propose a ferromagnet lattice magnetometer that coherently combines multiple levitated ferromagnets to enhance effective sensitivity. By replacing a single ferromagnet with a lattice, we increase the total polarized spin while preserving the intrinsic dynamical response of each constituent ferromagnet. We show that magnetic dipole-dipole interactions within the lattice can be dynamically suppressed through a high-frequency magnetic field, rendering the system effectively noninteracting, at the cost of only a moderate reduction in signal amplitude due to the distinct renormalization of linear and quadratic spin responses. We analyze the noise properties of the lattice and demonstrate that collective readout leads to favorable scaling with the number of ferromagnets. Interpreted in terms of axion-electron, dark photon, and axion-photon couplings, our results yield projected sensitivities that significantly exceed existing single-ferromagnet implementations. In particular, for axion-photon interactions, we find a nontrivial lattice-induced enhancement of the signal itself, leading to sensitivities that surpass existing constraints over a broad mass range.

hep-ph↗

arXiv2026

Thinking Seeds: Leveraging Historical Diversity for Position-Aware RL in LLMs

On-policy reinforcement learning (RL) for language model post-training suffers from a fundamental tension: as training progresses, policy entropy collapses and sampling diversity diminishes, causing the model to ``forget'' its own earlier exploratory capacity. While off-policy data can restore diversity, existing methods mix entire trajectories at the sequence level, introducing severe policy mismatch and training instability. We argue that the core question is not \emph{whether} to use off-policy data, but \emph{where} in the sequence it should appear. Based on this insight, we propose \textbf{Thinking Seeds}, a token-level mix-policy framework that uses the model's own historical checkpoints as off-policy prefixes, providing diverse starting points for reasoning, while the critical continuation is generated on-policy to preserve gradient quality. Through token-level importance ratios, Thinking Seeds effectively leverages historical diversity without compromising training stability. Extensive experiments across models and mathematical reasoning benchmarks demonstrate that Thinking Seeds consistently outperforms standard on-policy training and existing off-policy extensions. Our analysis reveals that the method maintains higher effective entropy, reduces gradient loss from clipping, and expands the explorable solution space, clarifying how position-aware mix-policy modeling improves both exploration and final performance in LLM RL.

cs.CL↗

arXiv2025

Efficient Training of Diffusion Mixture-of-Experts Models: A Practical Recipe

Recent efforts on Diffusion Mixture-of-Experts (MoE) models have primarily focused on developing more sophisticated routing mechanisms. However, we observe that the underlying architectural configuration space remains markedly under-explored. Inspired by the MoE design paradigms established in large language models (LLMs), we identify a set of crucial architectural factors for building effective Diffusion MoE models--including DeepSeek-style expert modules, alternative intermediate widths, varying expert counts, and enhanced attention positional encodings. Our systematic study reveals that carefully tuning these configurations is essential for unlocking the full potential of Diffusion MoE models, often yielding gains that exceed those achieved by routing innovations alone. Through extensive experiments, we present novel architectures that can be efficiently applied to both latent and pixel-space diffusion frameworks, which provide a practical and efficient training recipe that enables Diffusion MoE models to surpass strong baselines while using equal or fewer activated parameters. All code and models are publicly available at: https://github.com/yhlleo/EfficientMoE.

cs.LG↗

arXiv2025

Klear-AgentForge: Forging Agentic Intelligence through Posttraining Scaling

Despite the proliferation of powerful agentic models, the lack of critical post-training details hinders the development of strong counterparts in the open-source community. In this study, we present a comprehensive and fully open-source pipeline for training a high-performance agentic model for interacting with external tools and environments, named Klear-Qwen3-AgentForge, starting from the Qwen3-8B base model. We design effective supervised fine-tuning (SFT) with synthetic data followed by multi-turn reinforcement learning (RL) to unlock the potential for multiple diverse agentic tasks. We perform exclusive experiments on various agentic benchmarks in both tool use and coding domains. Klear-Qwen3-AgentForge-8B achieves state-of-the-art performance among LLMs of similar size and remains competitive with significantly larger models.

cs.AI↗

arXiv2025

SWIFT: A General Sensitive Weight Identification Framework for Fast Sensor-Transfer Pansharpening

Pansharpening aims to fuse high-resolution panchromatic (PAN) images with low-resolution multispectral (LRMS) images to generate high-resolution multispectral (HRMS) images. Although deep learning-based methods have achieved promising performance, they generally suffer from severe performance degradation when applied to data from unseen sensors. Adapting these models through full-scale retraining or designing more complex architectures is often prohibitively expensive and impractical for real-world deployment. To address this critical challenge, we propose a fast and general-purpose framework for cross-sensor adaptation, SWIFT (Sensitive Weight Identification for Fast Transfer). Specifically, SWIFT employs an unsupervised sampling strategy based on data manifold structures to balance sample selection while mitigating the bias of traditional Farthest Point Sampling, efficiently selecting only 3\% of the most informative samples from the target domain. This subset is then used to probe a source-domain pre-trained model by analyzing the gradient behavior of its parameters, allowing for the quick identification and subsequent update of only the weight subset most sensitive to the domain shift. As a plug-and-play framework, SWIFT can be applied to various existing pansharpening models. Extensive experiments demonstrate that SWIFT reduces the adaptation time from hours to approximately one minute on a single NVIDIA RTX 4090 GPU. The adapted models not only substantially outperform direct-transfer baselines but also achieve performance competitive with, and in some cases superior to, full retraining, establishing a new state-of-the-art on cross-sensor pansharpening tasks for the WorldView-2 and QuickBird datasets.

cs.CV↗

arXiv2025

Constraints on Axion Dark Matter by Spin-Dependent Macroscopic Force

Axion-like particles (ALPs) are hypothetical particles that serve as promising candidates for cold dark matter. Portals like inelastic axion scattering and axion propagated force have been employed to search for the upper limit of the ALPs' coupling with standard model particles. Other methods, like the axion-fermion interaction in the CASPEr experiment, integrate the dark matter motion into the measurement. We suggest a new method for detecting dark matter axions based on axionelectron elastic scattering. In the pseudoscalar axion model, this interaction can be seen as an effective magnetic field, so high-sensitivity atomic magnetometers can be utilized to measure this interaction. The scattering cross section of this process is significantly amplified by the high number density and occupation number of axion dark matter. The upper limit of the electron-axion coupling coefficient obtained from this process can reach two orders of magnitude higher than previous results at low axion mass, and will exceed the astrophysics limits by using a developing magnetometer. This scattering process also provides an efficient way to detect local structures of dark matter.

hep-ph↗

arXiv2025

Leanabell-Prover-V2: Verifier-integrated Reasoning for Formal Theorem Proving via Reinforcement Learning

We introduce our Leanabell-Prover-V2, a 7B large language models (LLMs) that can produce formal theorem proofs in Lean 4, with verifier-integrated Long Chain-of-Thoughts (CoT). Following our previous work Leanabell-Prover-V1, we continual to choose to posttrain existing strong prover models for further performance improvement. In our V2 version, we mainly upgrade the Reinforcement Learning (RL) with feedback provided by the Lean 4 verifier. Crucially, verifier feedback, such as indicating success or detailing specific errors, allows the LLM to become ``self-aware'' of the correctness of its own reasoning process and learn to reflexively correct errors. Leanabell-Prover-V2 directly optimizes LLM reasoning trajectories with multi-turn verifier interactions, together with feedback token masking for stable RL training and a simple reward strategy. Experiments show that Leanabell-Prover-V2 improves performance by 3.2% (pass@128) with Kimina-Prover-Preview-Distill-7B and 2.0% (pass@128) with DeepSeek-Prover-V2-7B on the MiniF2F test set. The source codes, curated data and models are available at: https://github.com/Leanabell-LM/Leanabell-Prover-V2.

cs.AI↗

arXiv2025

Evaluating Multimodal Large Language Models on Video Captioning via Monte Carlo Tree Search

Video captioning can be used to assess the video understanding capabilities of Multimodal Large Language Models (MLLMs). However, existing benchmarks and evaluation protocols suffer from crucial issues, such as inadequate or homogeneous creation of key points, exorbitant cost of data creation, and limited evaluation scopes. To address these issues, we propose an automatic framework, named AutoCaption, which leverages Monte Carlo Tree Search (MCTS) to construct numerous and diverse descriptive sentences (\textit{i.e.}, key points) that thoroughly represent video content in an iterative way. This iterative captioning strategy enables the continuous enhancement of video details such as actions, objects' attributes, environment details, etc. We apply AutoCaption to curate MCTS-VCB, a fine-grained video caption benchmark covering video details, thereby enabling a comprehensive evaluation of MLLMs on the video captioning task. We evaluate more than 20 open- and closed-source MLLMs of varying sizes on MCTS-VCB. Results show that MCTS-VCB can effectively and comprehensively evaluate the video captioning capability, with Gemini-1.5-Pro achieving the highest F1 score of 71.2. Interestingly, we fine-tune InternVL2.5-8B with the AutoCaption-generated data, which helps the model achieve an overall improvement of 25.0% on MCTS-VCB and 16.3% on DREAM-1K, further demonstrating the effectiveness of AutoCaption. The code and data are available at https://github.com/tjunlp-lab/MCTS-VCB.

cs.CV↗

arXiv2025

FF-PNet: A Pyramid Network Based on Feature and Field for Brain Image Registration

In recent years, deformable medical image registration techniques have made significant progress. However, existing models still lack efficiency in parallel extraction of coarse and fine-grained features. To address this, we construct a new pyramid registration network based on feature and deformation field (FF-PNet). For coarse-grained feature extraction, we design a Residual Feature Fusion Module (RFFM), for fine-grained image deformation, we propose a Residual Deformation Field Fusion Module (RDFFM). Through the parallel operation of these two modules, the model can effectively handle complex image deformations. It is worth emphasizing that the encoding stage of FF-PNet only employs traditional convolutional neural networks without any attention mechanisms or multilayer perceptrons, yet it still achieves remarkable improvements in registration accuracy, fully demonstrating the superior feature decoding capabilities of RFFM and RDFFM. We conducted extensive experiments on the LPBA and OASIS datasets. The results show our network consistently outperforms popular methods in metrics like the Dice Similarity Coefficient.

cs.CV↗

arXiv2025

Routing to the Right Expertise: A Trustworthy Judge for Instruction-based Image Editing

Instruction-based Image Editing (IIE) models have made significantly improvement due to the progress of multimodal large language models (MLLMs) and diffusion models, which can understand and reason about complex editing instructions. In addition to advancing current IIE models, accurately evaluating their output has become increasingly critical and challenging. Current IIE evaluation methods and their evaluation procedures often fall short of aligning with human judgment and often lack explainability. To address these limitations, we propose JUdgement through Routing of Expertise (JURE). Each expert in JURE is a pre-selected model assumed to be equipped with an atomic expertise that can provide useful feedback to judge output, and the router dynamically routes the evaluation task of a given instruction and its output to appropriate experts, aggregating their feedback into a final judge. JURE is trustworthy in two aspects. First, it can effortlessly provide explanations about its judge by examining the routed experts and their feedback. Second, experimental results demonstrate that JURE is reliable by achieving superior alignment with human judgments, setting a new standard for automated IIE evaluation. Moreover, JURE's flexible design is future-proof - modular experts can be seamlessly replaced or expanded to accommodate advancements in IIE, maintaining consistently high evaluation quality. Our evaluation data and results are available at https://github.com/Cyyyyyrus/JURE.git.

cs.AI↗