arXiv subjects

Lifeng Xing

Publications and source records attributed to Lifeng Xing.

4 recordsLinked to original sources

arXiv2026

LINGO: Latent Initialization and Gradient Optimization for Sparse-view X-ray Novel View Synthesis and CT Reconstruction with 3D Gaussian Splatting

In novel view synthesis and Computed Tomography (CT) reconstruction with sparse-view X-ray imaging, insufficient angular coverage leads to structural ambiguity and accumulated noise. Integrating 3D Gaussian Splatting (3DGS) with X-ray absorption physics can achieve promising results, but it suffers from noisy initialization, positional insensitivity, and weak gradients in low-density regions. In this paper, we propose a unified Latent Initialization and Gradient Optimization (LINGO) framework to address these issues. LINGO combines latent mask-space initialization with dynamic gradient optimization to improve point cloud structural completeness while accelerating training. It constructs voxel-level 3D filters from X-ray masks to robustly suppress background noise and provide reliable geometric priors. By employing an adaptive voxel scaling strategy and dynamically scaling loss, LINGO can adjust spatial resolution and explicitly amplify gradients in low-density structures. To evaluate the quality of initialization, we introduce the Initialization Point Cloud Structural Deviation (IPSD) metric. Experiments on the X3D dataset indicate that for the novel view synthesis task, LINGO improves the Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) by an average of 0.72 and 0.0039, respectively, over baselines under identical sparse-view settings, achieving comparable reconstruction quality within 5k steps to state-of-the-art models typically trained with 30k iterations. For the CT reconstruction task, LINGO also demonstrates consistent improvements, with average PSNR and SSIM gains of 0.36 and 0.0134. These results highlight LINGO's effectiveness in both accelerating training and enhancing reconstruction quality across different sparse-view imaging scenarios.

cs.CV↗

arXiv2025

RoNFA: Robust Neural Field-based Approach for Few-Shot Image Classification with Noisy Labels

In few-shot learning (FSL), the labeled samples are scarce. Thus, label errors can significantly reduce classification accuracy. Since label errors are inevitable in realistic learning tasks, improving the robustness of the model in the presence of label errors is critical. This paper proposes a new robust neural field-based image approach (RoNFA) for few-shot image classification with noisy labels. RoNFA consists of two neural fields for feature and category representation. They correspond to the feature space and category set. Each neuron in the field for category representation (FCR) has a receptive field (RF) on the field for feature representation (FFR) centered at the representative neuron for its category generated by soft clustering. In the prediction stage, the range of these receptive fields adapts according to the neuronal activation in FCR to ensure prediction accuracy. These learning strategies provide the proposed model with excellent few-shot learning capability and strong robustness against label noises. The experimental results on real-world FSL datasets with three different types of label noise demonstrate that the proposed method significantly outperforms state-of-the-art FSL methods. Its accuracy obtained in the presence of noisy labels even surpasses the results obtained by state-of-the-art FSL methods trained on clean support sets, indicating its strong robustness against noisy labels.

cs.CV↗

arXiv2023

MERGE: Fast Private Text Generation

The drastic increase in language models' parameters has led to a new trend of deploying models in cloud servers, raising growing concerns about private inference for Transformer-based models. Existing two-party privacy-preserving techniques, however, only take into account natural language understanding (NLU) scenarios. Private inference in natural language generation (NLG), crucial for applications like translation and code completion, remains underexplored.In addition, previous privacy-preserving techniques suffer from convergence issues during model training and exhibit poor inference speed when used with NLG models due to the neglect of time-consuming operations in auto-regressive generations. To address these issues, we propose a fast private text generation framework for Transformer-based language models, namely MERGE.MERGE reuses the output hidden state as the word embedding to bypass the embedding computation and reorganize the linear operations in the Transformer module to accelerate the forward procedure. Extensive experiments show that MERGE achieves a 26.5x speedup to the vanilla encrypted model under the sequence length 512, and reduces 80\% communication cost, with an up to 10x speedup to state-of-the-art approximated models.

cs.CL↗

arXiv2022

Mass-Ratio Distribution of Binaries From the LAMOST-MRS Survey

Binary evolution leads to the formation of important objects crucial to the development of astrophysics, but the statistical properties of binary populations are still poorly understood. The LAMOST-MRS has provided a large sample of stars to study the properties of binary populations, especially for the mass ratio distributions and the binary fractions. We have devised a Peak Amplitude Ratio (PAR) approach to derive the mass ratio of a binary system based on results obtained from its spectrum. By computing a cross-correlation function (CCF), we established a relationship between the derived mass ratio and the PARs of the binary systems. By utilizing spectral observations obtained from LAMSOT DR6 & DR7, we applied the PAR approach to form distributions of the derived mass ratio of the binary systems to the spectral types. We selected the mass ratio within the range of $0.6-1.0$ for investigating the mass-ratio distribution. Through a power-law fitting, we obtained the power index $γ$ values of $-0.42\pm0.27$, $0.03\pm0.12$, and $2.12\pm0.19$ for A-, F-, and G-type stars identified in the sample, respectively. The derived $γ$-values display an increasing trend toward lower primary star masses, and G-type binaries tend to be more in twins. The close binary fractions (for $P\lesssim 150\,{\rm d}$ and $q\gtrsim 0.6$) in our sample for A, F and G binaries are $7.6\pm 0.5 \%$, $4.9\pm 0.2 \%$ and $3.7 \pm 0.1 \%$, respectively. Note that the PAR approach can be applied to large spectroscopic surveys of stars.

astro-ph.SR↗