arXiv subjects

Qiuyang Zhang

Publications and source records attributed to Qiuyang Zhang.

5 recordsLinked to original sources

arXiv2026

Limited Reference, Reliable Generation: A Two-Component Framework for Tabular Data Generation in Low-Data Regimes

Synthetic tabular data generation is increasingly essential in machine learning, supporting downstream applications when real-world, high-quality tabular data is insufficient. Existing tabular generation approaches, such as generative adversarial networks (GANs) and fine-tuned Large Language Models (LLMs), typically require sufficient reference data, limiting their effectiveness in domain-specific datasets with scarce records. While prompt-based LLMs offer flexibility without parameter tuning, they often generate distributionally drifted data with localized redundancy, leading to degradation in downstream task performance. To overcome these issues, we propose ReFine, a framework that (i) extracts symbolic if-then rules from interpretable models and embeds them into prompts to explicitly guide the generation process toward the domain-specific distribution, and (ii) applies dual-granularity filtering that mitigates over-sampling patterns while preserving rare but informative samples to reduce localized redundancy. Extensive experiments on diverse benchmarks demonstrate that ReFine provides robust downstream utility, achieving a top-tier average rank across datasets and data regimes, with an average relative improvement of 7.48% in extreme low-data regimes.

cs.LG↗

arXiv2026

ScoutAttention: Efficient KV Cache Offloading via Layer-Ahead CPU Pre-computation for LLM Inference

Large language models encounter critical GPU memory capacity constraints during long-context inference, where KV cache memory consumption severely limits decode batch sizes. While existing research has explored offloading KV cache to DRAM, these approaches either demand frequent GPU-CPU data transfers or impose extensive CPU computation requirements, resulting in poor GPU utilization as the system waits for I/O operations or CPU processing to complete. We propose ScoutAttention, a novel KV cache offloading framework that accelerates LLM inference through collaborative GPU-CPU attention computation. To prevent CPU computation from bottlenecking the system, ScoutAttention introduces GPU-CPU collaborative block-wise sparse attention that significantly reduces CPU load. Unlike conventional parallel computing approaches, our framework features a novel layer-ahead CPU pre-computation algorithm, enabling the CPU to initiate attention computation one layer in advance, complemented by asynchronous periodic recall mechanisms to maintain minimal CPU compute load. Experimental results demonstrate that ScoutAttention maintains accuracy within 2.4% of baseline while achieving 2.1x speedup compared to existing offloading methods.

cs.LG↗

arXiv2026

SpikeTrack: A Spike-driven Framework for Efficient Visual Tracking

Spiking Neural Networks (SNNs) promise energy-efficient vision, but applying them to RGB visual tracking remains difficult: Existing SNN tracking frameworks either do not fully align with spike-driven computation or do not fully leverage neurons' spatiotemporal dynamics, leading to a trade-off between efficiency and accuracy. To address this, we introduce SpikeTrack, a spike-driven framework for energy-efficient RGB object tracking. SpikeTrack employs a novel asymmetric design that uses asymmetric timestep expansion and unidirectional information flow, harnessing spatiotemporal dynamics while cutting computation. To ensure effective unidirectional information transfer between branches, we design a memory-retrieval module inspired by neural inference mechanisms. This module recurrently queries a compact memory initialized by the template to retrieve target cues and sharpen target perception over time. Extensive experiments demonstrate that SpikeTrack achieves the state-of-the-art among SNN-based trackers and remains competitive with advanced ANN trackers. Notably, it surpasses TransT on LaSOT dataset while consuming only 1/26 of its energy. To our knowledge, SpikeTrack is the first spike-driven framework to make RGB tracking both accurate and energy efficient. The code and models are available at https://github.com/faicaiwawa/SpikeTrack.

cs.CV↗

arXiv2002

Time-Dependent Solution for a Star Immersed in a Background Radiation

We study a time-dependent and spherically-symmetric solution with a star-like source. We show that this solution can be interpreted as an exterior solution of a contracting star which has a decreasing temperature and is immersed in a homogenous and isotropic background radiation. Distribution of the temperature in the fields and close-to Schwarzschild approximation of the solution are studied. By identifying the radiation with the cosmic background one, we find that the close-to-Schwarzschild approximate solution is valid in a wide range in our solar system. Possible experimental tests of the solution are discussed briefly.

gr-qc↗

arXiv2002

Quantum Effects of an Extra Compact Dimension on the Wave Function of the Universe

We extend the direct quantum approach of the standard FRW cosmology from 4D to 5D and obtain a Hamiltonian formulation for a wave-like 5D FRW cosmology. Using a late-time approximation we isolate out a y-part from the full wave function of the 5D Universe. Then we find that the compactness of the fifth dimension y yields a quantized spectrum for the momentum $P_{5}$ along the fifth dimension, and we show that the whole space-part of the wave function of the 5D Universe satisfies a two-dimensional Schrödinger equation.

gr-qc↗