arXiv subjects

Wenli Li

Publications and source records attributed to Wenli Li.

10 recordsLinked to original sources

arXiv2026

Evolving from Lessons: Skill-Augmented Table Graph Reasoning for Operation-wise Table Question Answering

Table Question Answering (TableQA) aims to reason over tables to answer user queries. Existing research treats all questions uniformly and evaluates solely through overall accuracy, obscuring a critical reality that LLMs excel at simple lookups yet struggle with complex operations like aggregation and arithmetic. To reveal this disparity, we introduce a novel \emph{Operation-wise TableQA} task with a fine-grained question taxonomy and release two datasets named WikiTQ-ow and TabFact-ow for evaluation. As for modeling bottlenecks, existing methods flatten tables into linearized texts, disrupting inherent structures and inducing the ``lost-in-the-middle'' issue, which poses a primary barrier to complex cross-row reasoning. Moreover, they typically reason from scratch, neglecting reusable patterns shared across similar operations. To address these limitations, we propose a Skill-augmented Table Graph Reasoning (SkillTGR) framework for self-evolving structured reasoning. Specifically, SkillTGR represents tables as attributed graphs with explicit row-column-cell structures, where LLMs plan and execute dynamic chains to retrieve evidence subgraphs for graph traversal reasoning. Based on this, SkillTGR builds a hierarchical SkillBank to distill reason trajectories into abstract skills under cognitive heuristics, then hybrid retrieves both successful and failed skills for contrastive augmented table graph reasoning, thereby enabling the continual self-evolution. Extensive experiments demonstrate that SkillTGR achieves superior performance with an average of 5.91\% overall and 6.03\% operation-wise improvement, also reducing 19.76\% token consumption and 27.64\% inference latency. Our codes and data will be released upon publication.

cs.AI↗

arXiv2026

Cross-Domain Radiomap Prediction for Multi-Scatterer Environment: A Spherical-Wave-Based Approach

Radiomap prediction has found extensive applications in network planning and optimization. The radiomap is implicitly determined by electromagnetic (EM) wave propagation. The evolution of wireless communication toward higher frequency spectra has highlighted the effect of mesoscopic (i.e. wavelength-comparable scale) scatterers, which is negligible in the sub-6 GHz spectrum. To address this challenge, we develop a linear forward channel model to capture the propagation behavior in the multi-scatterer environment. The proposed model utilizes spherical-wave mode expansion to track the source radiation pattern, including scattering from a single scatterer and interactions among multiple scatterers. Both phenomena are represented as a superposition of spherical-wave modes, effectively capturing the multipath effect from a wave-based perspective. This forward model is then employed to formulate an inverse optimization problem, where the scattering responses and scatterer locations are jointly learned from sparse field measurements. Simulation results demonstrate that the proposed model accurately reconstructs and extrapolates the radiomap in both the spatial and the beam domain, further enables frequency-domain channel interpolation.

cs.IT↗

arXiv2026

SeGPruner: Semantic-Geometric Visual Token Pruner for 3D Question Answering

Vision-language models (VLMs) have been widely adopted for 3D question answering (3D QA). In typical pipelines, visual tokens extracted from multiple viewpoints are concatenated with language tokens and jointly processed by a large language model (LLM) for inference. However, aggregating multi-view observations inevitably introduces severe token redundancy, leading to an overly large visual token set that significantly hinders inference efficiency under constrained token budgets. Visual token pruning has emerged as a prevalent strategy to address this issue. Nevertheless, most existing pruners are primarily tailored to 2D inputs or rely on indirect geometric cues, which limits their ability to explicitly retain semantically critical objects and maintain sufficient spatial coverage for robust 3D reasoning. In this paper, we propose SeGPruner, a semantic-aware and geometry-guided token reduction framework for efficient 3D QA with multi-view images. Specifically, SeGPruner first preserves semantically salient tokens through an attention-based importance module (Saliency-aware Token Selector), ensuring that object-critical evidence is retained. It then complements these tokens with spatially diverse ones via a geometry-guided selector (Geometry-aware Token Diversifier), which jointly considers semantic relevance and 3D geometric distance. This cooperation between saliency preservation and geometry-guided diversification balances object-level evidence and global scene coverage under aggressive token reduction. Extensive experiments on ScanQA and OpenEQA demonstrate that SeGPruner substantially improves inference efficiency, reducing the visual token budget by 91% and inference latency by 86%, while maintaining competitive performance in 3D reasoning tasks.

cs.CV↗

arXiv2026

Towards Any-Quality Image Segmentation via Generative and Adaptive Latent Space Enhancement

Segment Anything Models (SAMs), known for their exceptional zero-shot segmentation performance, have garnered significant attention in the research community. Nevertheless, their performance drops significantly on severely degraded, low-quality images, limiting their effectiveness in real-world scenarios. To address this, we propose GleSAM++, which utilizes Generative Latent space Enhancement to boost robustness on low-quality images, thus enabling generalization across various image qualities. Additionally, to improve compatibility between the pre-trained diffusion model and the segmentation framework, we introduce two techniques, i.e., Feature Distribution Alignment (FDA) and Channel Replication and Expansion (CRE). However, the above components lack explicit guidance regarding the degree of degradation. The model is forced to implicitly fit a complex noise distribution that spans conditions from mild noise to severe artifacts, which substantially increases the learning burden and leads to suboptimal reconstructions. To address this issue, we further introduce a Degradation-aware Adaptive Enhancement (DAE) mechanism. The key principle of DAE is to decouple the reconstruction process for arbitrary-quality features into two stages: degradation-level prediction and degradation-aware reconstruction. Our method can be applied to pre-trained SAM and SAM2 with only minimal additional learnable parameters, allowing for efficient optimization. Extensive experiments demonstrate that GleSAM++ significantly improves segmentation robustness on complex degradations while maintaining generalization to clear images. Furthermore, GleSAM++ also performs well on unseen degradations, underscoring the versatility of our approach and dataset.

cs.CV↗

arXiv2025

MGTS-Net: Exploring Graph-Enhanced Multimodal Fusion for Augmented Time Series Forecasting

Recent research in time series forecasting has explored integrating multimodal features into models to improve accuracy. However, the accuracy of such methods is constrained by three key challenges: inadequate extraction of fine-grained temporal patterns, suboptimal integration of multimodal information, and limited adaptability to dynamic multi-scale features. To address these problems, we propose MGTS-Net, a Multimodal Graph-enhanced Network for Time Series forecasting. The model consists of three core components: (1) a Multimodal Feature Extraction layer (MFE), which optimizes feature encoders according to the characteristics of temporal, visual, and textual modalities to extract temporal features of fine-grained patterns; (2) a Multimodal Feature Fusion layer (MFF), which constructs a heterogeneous graph to model intra-modal temporal dependencies and cross-modal alignment relationships and dynamically aggregates multimodal knowledge; (3) a Multi-Scale Prediction layer (MSP), which adapts to multi-scale features by dynamically weighting and fusing the outputs of short-term, medium-term, and long-term predictors. Extensive experiments demonstrate that MGTS-Net exhibits excellent performance with light weight and high efficiency. Compared with other state-of-the-art baseline models, our method achieves superior performance, validating the superiority of the proposed methodology.

cs.LG↗

arXiv2025

Format as a Prior: Quantifying and Analyzing Bias in LLMs for Heterogeneous Data

Large Language Models (LLMs) are increasingly employed in applications that require processing information from heterogeneous formats, including texts, tables, infoboxes, and knowledge graphs. However, systematic biases toward particular formats may undermine LLMs' ability to integrate heterogeneous data impartially, potentially resulting in reasoning errors and increased risks in downstream tasks. Yet it remains unclear whether such biases are systematic, which data-level factors drive them, and what internal mechanisms underlie their emergence. In this paper, we present the first comprehensive study of format bias in LLMs through a three-stage empirical analysis. The first stage explores the presence and direction of bias across a diverse range of LLMs. The second stage examines how key data-level factors influence these biases. The third stage analyzes how format bias emerges within LLMs' attention patterns and evaluates a lightweight intervention to test its effectiveness. Our results show that format bias is consistent across model families, driven by information richness, structure quality, and representation type, and is closely associated with attention imbalance within the LLMs. Based on these investigations, we identify three future research directions to reduce format bias: enhancing data pre-processing through format repair and normalization, introducing inference-time interventions such as attention re-weighting, and developing format-balanced training corpora. These directions will support the design of more robust and fair heterogeneous data processing systems.

cs.CL↗

arXiv2022

Super-resolution multicolor fluorescence microscopy enabled by an apochromatic super-oscillatory lens with extended depth-of-focus

Multicolor super-resolution imaging remains an intractable challenge for both far-field and near-field based super-resolution techniques. Planar super-oscillatory lens (SOL), a far-field subwavelength-focusing diffractive lens device, holds great potential for achieving sub-diffraction-limit imaging at multiple wavelengths. However, conventional SOL devices suffer from a numerical aperture (NA) related intrinsic tradeoff among the depth of focus (DoF), chromatic dispersion and focus spot size, being an essential characteristics of common diffractive optical elements. Typically, the limited DoF and significant chromatism associated with high NA can lead to unfavorable degradation of image quality although increasing NA imporves the resolution. Here, we apply a multi-objective genetic algorithm (GA) optimization approach to design an apochromatic binary-phase SOL that generates axially jointed multifoci concurrently having prolonged DoF, customized working distance (WD) and suppressed side-lobes yet minimized main-lobe size, optimizing the aforementioned NA-dependent tradeoff. Experimental implementation of this GA-optimized SOL demonstrates simultaneous focusing of blue, green and red light beams into an optical needle half of the incident wavelength in diameter at 428 um WD, resulting in an ultimate resolution better than one third of the incident wavelength in the lateral dimension. By integrating this apochromatic SOL device with a commercial fluorescence microscope, we employ the optical needle to perform, for the first time, three-dimensional super-resolution multicolor fluorescence imaging of the unseen fine structure of neurons at one go. The present study provides not only a practical route to far-field multicolor super-resolution imaging but also a viable approach for constructing imaging systems avoiding complex sample positioning and unfavorable photobleaching.

physics.optics↗

arXiv2021

Preparation and Fluorescent Wavelength Control of Multi-Color Nitrogen-Doped Carbon Nano-Dots

It is known that, by taking advantage of heteroatom doping, the electronic states and transition channels in carbon nano-dots (CNDs) can be effectively modulated. Thus, the photoluminescence (PL) properties of CNDs can be changed. For potential applications of CNDs as advanced materials for optoelectronic devices, it is important and significant to develop the practical techniques for doping heteroatoms into CNDs. In this work, we synthesize the multi-color fluorescent by using a fast and effective microwave method where the CNDs are nitrogen-doped. We examine the influence of different ratios of the raw materials on the structure and optical properties of N-CNDs. The results show that the prepared N-CNDs can generate blue (445 nm), green (546 nm), and orange (617 nm) fluorescence or PL with the mass ratio of the raw materials at 1:1, 1:2 and 1:3, respectively. We find that the N content in N-CNDs leads to different surface/edge states in $n-π^*$ domain. Thus, the wavelength of the PL emission from N-CNDs can be tuned via controlling the N concentration doped into the CNDs. Moreover, it is shown that the intensity of the PL from N-CNDs is mainly positively related to the content of C-O groups attached on the surface/edges of the N-CNDs. This study provides an effective experimental method and technical way to improve the fluorescent emission, and to modulate the color of the PL emission from CNDs.

cond-mat.mes-hall↗

arXiv2020

Siting thousands of radio transmitter towers on terrains with billions of points

This paper presents a system that sites (finds optimal locations for) thousands of radio transmitter towers on terrains of up to two billion elevation posts. Applications include cellphone towers, camera systems, or even mitigating environmental visual nuisances. The transmitters and receivers may be situated above the terrain. The system has been parallelized with OpenMP to run on a multicore CPU.

eess.SP↗

arXiv2017

Hausdorff measure of escaping sets on certain meromorphic functions

We consider transcendental meromorphic function for which the set of finite singularities of its inverse is bounded. Bergweiler and Kotus gave bounds for the Hausdorff dimension of escaping sets if the function has no logarithmic singularities over infinity, the multiplicities of poles are bounded and the order is finite. We study the case of infinite order and find gauge functions for which the Hausdorff measure of escaping sets is zero or infinity.

math.DS↗