arXiv subjects

Fei Li

Publications and source records attributed to Fei Li.

At least 55 records · Page 3Linked to original sources

arXiv2026

An Efficient Additive Kolmogorov-Arnold Transformer for Point-Level Maize Localization in Unmanned Aerial Vehicle Imagery

High-resolution UAV photogrammetry has become a key technology for precision agriculture, enabling centimeter-level crop monitoring and point-level plant localization. However, point-level maize localization in UAV imagery remains challenging due to (1) extremely small object-to-pixel ratios, typically less than 0.1%, (2) prohibitive computational costs of quadratic attention on ultra-high-resolution images larger than 3000 x 4000 pixels, and (3) agricultural scene-specific complexities such as sparse object distribution and environmental variability that are poorly handled by general-purpose vision models. To address these challenges, we propose the Additive Kolmogorov-Arnold Transformer (AKT), which replaces conventional multilayer perceptrons with Pade Kolmogorov-Arnold Network (PKAN) modules to enhance functional expressivity for small-object feature extraction, and introduces PKAN Additive Attention (PAA) to model multiscale spatial dependencies with reduced computational complexity. In addition, we present the Point-based Maize Localization (PML) dataset, consisting of 1,928 high-resolution UAV images with approximately 501,000 point annotations collected under real field conditions. Extensive experiments show that AKT achieves an average F1-score of 62.8%, outperforming state-of-the-art methods by 4.2%, while reducing FLOPs by 12.6% and improving inference throughput by 20.7%. For downstream tasks, AKT attains a mean absolute error of 7.1 in stand counting and a root mean square error of 1.95-1.97 cm in interplant spacing estimation. These results demonstrate that integrating Kolmogorov-Arnold representation theory with efficient attention mechanisms offers an effective framework for high-resolution agricultural remote sensing.

cs.CV↗

arXiv2025

VSA:Visual-Structural Alignment for UI-to-Code

The automation of user interface development has the potential to accelerate software delivery by mitigating intensive manual implementation. Despite the advancements in Large Multimodal Models for design-to-code translation, existing methodologies predominantly yield unstructured, flat codebases that lack compatibility with component-oriented libraries such as React or Angular. Such outputs typically exhibit low cohesion and high coupling, complicating long-term maintenance. In this paper, we propose \textbf{VSA (VSA)}, a multi-stage paradigm designed to synthesize organized frontend assets through visual-structural alignment. Our approach first employs a spatial-aware transformer to reconstruct the visual input into a hierarchical tree representation. Moving beyond basic layout extraction, we integrate an algorithmic pattern-matching layer to identify recurring UI motifs and encapsulate them into modular templates. These templates are then processed via a schema-driven synthesis engine, ensuring the Large Language Model generates type-safe, prop-drilled components suitable for production environments. Experimental results indicate that our framework yields a substantial improvement in code modularity and architectural consistency over state-of-the-art benchmarks, effectively bridging the gap between raw pixels and scalable software engineering.

cs.IR↗

arXiv2025

Modular Layout Synthesis (MLS): Front-end Code via Structure Normalization and Constrained Generation

Automated front-end engineering drastically reduces development cycles and minimizes manual coding overhead. While Generative AI has shown promise in translating designs to code, current solutions often produce monolithic scripts, failing to natively support modern ecosystems like React, Vue, or Angular. Furthermore, the generated code frequently suffers from poor modularity, making it difficult to maintain. To bridge this gap, we introduce Modular Layout Synthesis (MLS), a hierarchical framework that merges visual understanding with structural normalization. Initially, a visual-semantic encoder maps the screen capture into a serialized tree topology, capturing the essential layout hierarchy. Instead of simple parsing, we apply heuristic deduplication and pattern recognition to isolate reusable blocks, creating a framework-agnostic schema. Finally, a constraint-based generation protocol guides the LLM to synthesize production-ready code with strict typing and component props. Evaluations show that MLS significantly outperforms existing baselines, ensuring superior code reusability and structural integrity across multiple frameworks

cs.IR↗

arXiv2025

Event Extraction in Large Language Model

Large language models (LLMs) and multimodal LLMs are changing event extraction (EE): prompting and generation can often produce structured outputs in zero shot or few shot settings. Yet LLM based pipelines face deployment gaps, including hallucinations under weak constraints, fragile temporal and causal linking over long contexts and across documents, and limited long horizon knowledge management within a bounded context window. We argue that EE should be viewed as a system component that provides a cognitive scaffold for LLM centered solutions. Event schemas and slot constraints create interfaces for grounding and verification; event centric structures act as controlled intermediate representations for stepwise reasoning; event links support relation aware retrieval with graph based RAG; and event stores offer updatable episodic and agent memory beyond the context window. This survey covers EE in text and multimodal settings, organizing tasks and taxonomy, tracing method evolution from rule based and neural models to instruction driven and generative frameworks, and summarizing formulations, decoding strategies, architectures, representations, datasets, and evaluation. We also review cross lingual, low resource, and domain specific settings, and highlight open challenges and future directions for reliable event centric systems. Finally, we outline open challenges and future directions that are central to the LLM era, aiming to evolve EE from static extraction into a structurally reliable, agent ready perception and memory layer for open world systems.

cs.CL↗

arXiv2025

Efficient Redundancy Reduction for Open-Vocabulary Semantic Segmentation

Open-vocabulary semantic segmentation (OVSS) is an open-world task that aims to assign each pixel within an image to a specific class defined by arbitrary text descriptions. While large-scale vision-language models have shown remarkable open-vocabulary capabilities, their image-level pretraining limits effectiveness on pixel-wise dense prediction tasks like OVSS. Recent cost-based methods narrow this granularity gap by constructing pixel-text cost maps and refining them via cost aggregation mechanisms. Despite achieving promising performance, these approaches suffer from high computational costs and long inference latency. In this paper, we identify two major sources of redundancy in the cost-based OVSS framework: redundant information introduced during cost maps construction and inefficient sequence modeling in cost aggregation. To address these issues, we propose ERR-Seg, an efficient architecture that incorporates Redundancy-Reduced Hierarchical Cost maps (RRHC) and Redundancy-Reduced Cost Aggregation (RRCA). Specifically, RRHC reduces redundant class channels by customizing a compact class vocabulary for each image and integrates hierarchical cost maps to enrich semantic representation. RRCA alleviates computational burden by performing both spatial-level and class-level sequence reduction before aggregation. Overall, ERR-Seg results in a lightweight structure for OVSS, characterized by substantial memory and computational savings without compromising accuracy. Compared to previous state-of-the-art methods on the ADE20K-847 benchmark, ERR-Seg improves performance by $5.6\%$ while achieving a 3.1$\times$ speedup.

cs.CV↗

arXiv2025

PaSE: Prototype-aligned Calibration and Shapley-based Equilibrium for Multimodal Sentiment Analysis

Multimodal Sentiment Analysis (MSA) seeks to understand human emotions by integrating textual, acoustic, and visual signals. Although multimodal fusion is designed to leverage cross-modal complementarity, real-world scenarios often exhibit modality competition: dominant modalities tend to overshadow weaker ones, leading to suboptimal performance. In this paper, we propose PaSE, a novel Prototype-aligned Calibration and Shapley-optimized Equilibrium framework, which enhances collaboration while explicitly mitigating modality competition. PaSE first applies Prototype-guided Calibration Learning (PCL) to refine unimodal representations and align them through an Entropic Optimal Transport mechanism that ensures semantic consistency. To further stabilize optimization, we introduce a Dual-Phase Optimization strategy. A prototype-gated fusion module is first used to extract shared representations, followed by Shapley-based Gradient Modulation (SGM), which adaptively adjusts gradients according to the contribution of each modality. Extensive experiments on IEMOCAP, MOSI, and MOSEI confirm that PaSE achieves the superior performance and effectively alleviates modality competition.

cs.LG↗

arXiv2025

Generating Attribute-Aware Human Motions from Textual Prompt

Text-driven human motion generation has recently attracted considerable attention, allowing models to generate human motions based on textual descriptions. However, current methods neglect the influence of human attributes-such as age, gender, weight, and height-which are key factors shaping human motion patterns. This work represents a pilot exploration for bridging this gap. We conceptualize each motion as comprising both attribute information and action semantics, where textual descriptions align exclusively with action semantics. To achieve this, a new framework inspired by Structural Causal Models is proposed to decouple action semantics from human attributes, enabling text-to-semantics prediction and attribute-controlled generation. The resulting model is capable of generating attribute-aware motion aligned with the user's text and attribute inputs. For evaluation, we introduce a comprehensive dataset containing attribute annotations for text-motion pairs, setting the first benchmark for attribute-aware motion generation. Extensive experiments validate our model's effectiveness.

cs.CV↗

arXiv2025

Natural Realization of Tens-of-GeV Dark Matter in the GNMSSM

This study presents a comparative analysis of the Minimal Supersymmetric Standard Model (MSSM), the $Z_3$-symmetric Next-to-Minimal Supersymmetric Standard Model ($Z_3$-NMSSM), and the General Next-to-Minimal Supersymmetric Standard Model (GNMSSM), incorporating constraints from dark matter (DM) relic density, the LUX-ZEPLIN 2024 experiment (LZ 2024), Higgs data, and the Large Hadron Collider (LHC). The results suggest that, among the three frameworks, only GNMSSM can naturally accommodate for light DM with a mass below $100~{\rm GeV}$. As such, the viable supersymmetry candidate is primarily Singlino-like. One key advantage of the GNMSSM is the effective decoupling between interactions that establish the relic density and those that control direct detection, allowing the model to satisfy all current experimental bounds simultaneously. We further explore two characteristic mass hierarchies in the GNMSSM parameter space, each exhibiting distinct phenomenological behaviors. The first hierarchy, $\tilde{S} < \tilde{B} < \tilde{H}$ (Singlino--Bino--Higgsino), involves a relatively light Bino and allows the Higgsino mass parameter, $μ_{\rm tot}$, to be as low as about $200~{\rm GeV}$, naturally yielding light DM at tens of GeV. The dominant annihilation channels are then $\tildeχ_1^0\tildeχ_1^0 \to A_sA_s$ in the $h_1$ scenario and $\tildeχ_1^0\tildeχ_1^0 \to h_sA_s$ in the $h_2$ scenario, where $h_s$ and $A_s$ denote singlet-dominated CP-even and CP-odd Higgs bosons, respectively. The second hierarchy, $\tilde{S} < \tilde{H} < \tilde{B}$, corresponds to a heavy Bino. In this case, although the DM phenomenology remains qualitatively similar, LHC constraints require $μ_{\rm tot} \gtrsim 900~{\rm GeV}$, implying a significant degree of fine-tuning in reproducing the $Z$-boson mass.

hep-ph↗

arXiv2025

Channel Superposition Mitigates Photon Loss Errors in Quantum Illumination

In quantum illumination, the probe photon is entangled with an ancilla photon, and both are jointly measured at the end. The entanglement between the probe and ancilla photons enhances the detection performance per unit average photon number in the probe mode, particularly in low-reflectivity and high-noise scenarios. However, photon loss severely limits the practical advantage of such protocols. To address this, we employ a channel superposition framework, which encompasses two kinds of channel superposition protocols: indefinite causal order (ICO) and path superposition with disjoint environment (PS-DE). Our analytical and numerical analysis based on the quantum Chernoff bound shows that both ICO and PS-DE can, in principle, achieve an advantage. The advantage persists as long as non-zero interference remains, reverting to the performance of standard quantum illumination once the interference is completely suppressed. Crucially, the ICO protocol is significantly more robust, maintaining a tighter upper bound on the error probability than standard quantum illumination and the PS-DE approach. This performance hierarchy is rooted in their fundamental structures: ICO exploits a shared environment to generate stronger quantum interference, while PS-DE, relying on disjoint environments, offers a more experimentally tractable albeit less potent alternative.

quant-ph↗

arXiv2025

Attractive features of Higgsino Dark Matter in the Next-to-Minimal Supersymmetric Standard Model

In the Higgsino dark matter (DM) scenario of the Minimal Supersymmetric Model (MSSM), the mixing of Gaugino and Higgsino influences the mass splitting between neutralinos predominantly composed of Higgsino and introduces coupling between the DM and Higgs bosons. These effects modify the DM-nucleon scattering cross-section, causing conflicts with the latest direct detection results from LZ experiments for both substantial and minute mixings. Consequently, the experimental measurement of DM relic density necessitates the Higgsino DM mass to be approximately 1.1 TeV. We discovered that in the Higgsino DM scenario of the Next-to-Minimal Supersymmetric Model (NMSSM), the mixing of Higgsino and Singlino introduces analogous effects, with a crucial distinction being that the current LZ experiment permits significant mixing between Singlino and Higgsino. This pronounced mixing effect effectively attenuates the interactions between Higgsino-dominated neutralinos and standard model particles, enabling DM masses exceeding roughly 670 GeV to achieve the correct relic abundance. Through analytical formulas and numerical results, we elucidated these characteristics which were not observed before. Our research reveals that in the NMSSM, when comprehensively examining the mixing effects of Higgsino, Gaugino, and Singlino, the properties of Higgsino DM become markedly more intricate compared to the MSSM predictions.

hep-ph↗

arXiv2025

DiscoSG: Towards Discourse-Level Text Scene Graph Parsing through Iterative Graph Refinement

Vision-Language Models (VLMs) generate discourse-level, multi-sentence visual descriptions, challenging text scene graph parsers built for single-sentence caption-to-graph mapping. Current approaches typically merge sentence-level parsing outputs for discourse input, often missing phenomena like cross-sentence coreference, resulting in fragmented graphs and degraded downstream VLM task performance. We introduce a new task, Discourse-level text Scene Graph parsing (DiscoSG), and release DiscoSG-DS, a dataset of 400 expert-annotated and 8,430 synthesised multi-sentence caption-graph pairs. Each caption averages 9 sentences, and each graph contains at least 3 times more triples than those in existing datasets. Fine-tuning GPT-4o on DiscoSG-DS yields over 40% higher SPICE metric than the best sentence-merging baseline. However, its high inference cost and licensing restrict open-source use. Smaller fine-tuned open-source models (e.g., Flan-T5) perform well on simpler graphs yet degrade on denser, more complex graphs. To bridge this gap, we introduce DiscoSG-Refiner, a lightweight open-source parser that drafts a seed graph and iteratively refines it with a novel learned graph-editing model, achieving 30% higher SPICE than the baseline while delivering 86 times faster inference than GPT-4o. It generalises from simple to dense graphs, thereby consistently improving downstream VLM tasks, including discourse-level caption evaluation and hallucination detection, outperforming alternative open-source parsers. Code and data are available at https://github.com/ShaoqLin/DiscoSG .

cs.CL↗

arXiv2025

A Quantum-Inspired Algorithm for Solving Sudoku Puzzles and the MaxCut Problem

We propose and evaluate a quantum-inspired algorithm for solving Quadratic Unconstrained Binary Optimization (QUBO) problems, which are mathematically equivalent to finding ground states of Ising spin-glass Hamiltonians. The algorithm employs Matrix Product States (MPS) to compactly represent large superpositions of spin configurations and utilizes a discrete driving schedule to guide the MPS toward the ground state. At each step, a driver Hamiltonian -- incorporating a transverse magnetic field -- is combined with the problem Hamiltonian to enable spin flips and facilitate quantum tunneling. The MPS is updated using the standard Density Matrix Renormalization Group (DMRG) method, which iteratively minimizes the system's energy via multiple sweeps across the spin chain. Despite its heuristic nature, the algorithm reliably identifies global minima, not merely near-optimal solutions, across diverse QUBO instances. We first demonstrate its effectiveness on intermediate-level Sudoku puzzles from publicly available sources, involving over $200$ Ising spins with long-range couplings dictated by constraint satisfaction. We then apply the algorithm to MaxCut problems from the Biq Mac library, successfully solving instances with up to $251$ nodes and $3,265$ edges. We discuss the advantages of this quantum-inspired approach, including its scalability, generalizability, and suitability for industrial-scale QUBO applications.

cs.AI↗

arXiv2025

Cross-Document Cross-Lingual NLI via RST-Enhanced Graph Fusion and Interpretability Prediction

Natural Language Inference (NLI) is a fundamental task in natural language processing. While NLI has developed many sub-directions such as sentence-level NLI, document-level NLI and cross-lingual NLI, Cross-Document Cross-Lingual NLI (CDCL-NLI) remains largely unexplored. In this paper, we propose a novel paradigm: CDCL-NLI, which extends traditional NLI capabilities to multi-document, multilingual scenarios. To support this task, we construct a high-quality CDCL-NLI dataset including 25,410 instances and spanning 26 languages. To address the limitations of previous methods on CDCL-NLI task, we further propose an innovative method that integrates RST-enhanced graph fusion with interpretability-aware prediction. Our approach leverages RST (Rhetorical Structure Theory) within heterogeneous graph neural networks for cross-document context modeling, and employs a structure-aware semantic alignment based on lexical chains for cross-lingual understanding. For NLI interpretability, we develop an EDU (Elementary Discourse Unit)-level attribution framework that produces extractive explanations. Extensive experiments demonstrate our approach's superior performance, achieving significant improvements over both conventional NLI models as well as large language models. Our work sheds light on the study of NLI and will bring research interest on cross-document cross-lingual context understanding, hallucination elimination and interpretability inference. Our code and datasets are available at "https://github.com/Leonardo123-ui/CDCL_NLI" for peer review.

cs.CL↗

arXiv2025

Dense gas properties and star formation in M 82

High-density molecular gas plays a vital role in supporting star formation within galaxies. However, traditional tracers of dense gas, such as the low-$J$ transitions of HCN and HCO$^+$, are predominantly optically thick. This characteristic presents a significant challenge in accurately estimating the column density of dense gas and mapping its spatial distribution. Optically-thin tracers, including HC$_3$N (10-9) and isotopologues of HCN, HCO$^+$, HNC (1-0), among others, emerge as better tracers, by enabling more precise measurements and analyses of the dense gas in galaxies. Here, we present the first high-resolution ($\sim$3.8$''$, corresponding to $\sim$65 pc) molecular line observations of the nearby starburst galaxy M~82 with IRAM Northern Extended Millimeter Array (NOEMA). Notably, HC$_3$N (10-9) and H$^{13}$CN (1-0) emission lines are brighter at the northeast part (NE lobe) than those at the southwest part (SW lobe) of M~82, suggesting a higher accumulation of dense gas at the NE lobe. The spatial distributions of H41$α$ and 3-mm continuum emission indicate that starburst activity varies across the central starburst disk, with more intense star formation occurring at the SW lobe compared to the NE lobe. The average optical depths of HCN and HCO$^+$ (1-0) across the triple-peaked regions exhibit significant variation, with the highest values observed at the NE lobe. Our results suggest a potential evolutionary sequence in M~82, where the distributions of star formation and dense gas appear to be partially decoupled -- a phenomenon that classical dense gas tracers cannot adequately probe.

astro-ph.GA↗

arXiv2025

A transformer-BiGRU-based framework with data augmentation and confident learning for network intrusion detection

In today's fast-paced digital communication, the surge in network traffic data and frequency demands robust and precise network intrusion solutions. Conventional machine learning methods struggle to grapple with complex patterns within the vast network intrusion datasets, which suffer from data scarcity and class imbalance. As a result, we have integrated machine learning and deep learning techniques within the network intrusion detection system to bridge this gap. This study has developed TrailGate, a novel framework that combines machine learning and deep learning techniques. By integrating Transformer and Bidirectional Gated Recurrent Unit (BiGRU) architectures with advanced feature selection strategies and supplemented by data augmentation techniques, TrailGate can identifies common attack types and excels at detecting and mitigating emerging threats. This algorithmic fusion excels at detecting common and well-understood attack types and has the unique ability to swiftly identify and neutralize emerging threats that stem from existing paradigms.

cs.LG↗

arXiv2025

Baichuan-M2: Scaling Medical Capability with Large Verifier System

As large language models (LLMs) advance in conversational and reasoning capabilities, their practical application in healthcare has become a critical research focus. However, there is a notable gap between the performance of medical LLMs on static benchmarks such as USMLE and their utility in real-world clinical decision-making. This discrepancy arises because traditional exams fail to capture the dynamic, interactive nature of medical consultations. To address this challenge, we introduce a novel dynamic verification framework that moves beyond static answer verifier, establishing a large-scale, high-fidelity interactive reinforcement learning system. Our framework comprises two key components: a Patient Simulator that creates realistic clinical environments using de-identified medical records, and a Clinical Rubrics Generator that dynamically produces multi-dimensional evaluation metrics. Building on this foundation, we develop Baichuan-M2, a 32B-parameter medical augmented reasoning model trained through a multi-stage reinforcement learning strategy with an improved Group Relative Policy Optimization (GRPO) algorithm. Evaluated on HealthBench, Baichuan-M2 outperforms all other open-source models and most advanced closed-source counterparts, achieving a score above 32 on the challenging HealthBench Hard benchmark-previously exceeded only by GPT-5. Our work demonstrates that robust dynamic verifier system is essential for aligning LLM capabilities with practical clinical applications, establishing a new Pareto front in the performance-parameter trade-off for medical AI deployment.

cs.LG↗

arXiv2025

Cauchy matrix structures and solutions to the nonisospectral three-component mKdV equations

Nonisospectral integrable systems can describe solitary waves in nonuniform media. In this paper, we apply the Cauchy matrix approach to construct three types of nonisospectral matrix modified Korteweg-de Vries (mKdV) eqautions and present their Cauchy matrix structures and solutions. Further, through complex reduction, we further obtain three nonisospectral three-component mKdV (NTCmKdV) equations, which can be regarded as novel members of the nonisospectral Ablowitz-Kaup-Newell-Segur (AKNS) hierarchy. In particular, the explicit solutions are given for the soliton solutions, and the double-pole solutions, respectively. The dynamical behaviors of these solutions are analyzed to reveal the influence of nonisospectral terms on the solution structure.

nlin.SI↗

arXiv2025

Scaling Laws in Plasma Channels for Laser Wakefield Accelerators

Preformed plasma channels are essential for guiding high-power laser pulses over extended distances in laser wakefield accelerators, enabling the generation of multi-GeV electron beams for applications such as free-electron lasers and particle colliders. Above-threshold ionization heating provides a robust mechanism for creating laser-matched plasma channels across a wide parameter range, owing to its density- and geometry-independent heating effect. Establishing predictive scaling laws between channel parameters and formation conditions is critical for designing channels optimized for electron acceleration across energies spanning hundreds of MeV to tens of GeV. Through combined timescale analysis and numerical simulations, hydrodynamic expansion is identified as the dominant mechanism governing density profile evolution during ATI channel formation. Remarkably, this process maintains effective laser-guiding channel structures across a wide range of initial gas density, as evidenced by the persistent profile similarity observed despite these significant parameter variations. For parabolic channels matched to Gaussian laser drivers, rigorous scaling laws are established that, the on-axis density scales linearly with the initial gas density, while the matching radius has an exponential dependence on both the initial gas density and the ionization laser radius. These findings provide a systematic framework for the predictive design and optimization of plasma channels in high-efficiency and high-energy LWFA applications.

physics.plasm-ph↗