arXiv subjects

Qinyu Xu

Publications and source records attributed to Qinyu Xu.

5 recordsLinked to original sources

arXiv2025

Data-Driven Parameter Identification for Tumor Growth Models

Modeling tumor growth accurately is essential for understanding cancer progression and informing treatment strategies. To estimate the parameters in the tumor growth model described by a nonlinear PDE, we adopt Physics-Informed Neural Networks (PINNs) and DeepONet, which show advantages especially when the observation data is scarce and contains noise. With the help of real-life lab data, we have demonstrated the potential of applying deep learning tools to address data-driven modeling for tumor growth in biology.

math.AP↗

arXiv2025

High-order Interactions Modeling for Interpretable Multi-Agent Q-Learning

The ability to model interactions among agents is crucial for effective coordination and understanding their cooperation mechanisms in multi-agent reinforcement learning (MARL). However, previous efforts to model high-order interactions have been primarily hindered by the combinatorial explosion or the opaque nature of their black-box network structures. In this paper, we propose a novel value decomposition framework, called Continued Fraction Q-Learning (QCoFr), which can flexibly capture arbitrary-order agent interactions with only linear complexity $\mathcal{O}\left({n}\right)$ in the number of agents, thus avoiding the combinatorial explosion when modeling rich cooperation. Furthermore, we introduce the variational information bottleneck to extract latent information for estimating credits. This latent information helps agents filter out noisy interactions, thereby significantly enhancing both cooperation and interpretability. Extensive experiments demonstrate that QCoFr not only consistently achieves better performance but also provides interpretability that aligns with our theoretical analysis.

cs.MA↗

arXiv2025

Parameter Identifiability of RNA Dynamics in PDE Transport Models of Fluorescence Recovery After Photobleaching

The transport and localization of RNA molecules, crucial for cellular function and development, involve a combination of diffusion and active transport mechanisms. Here, we are motivated by understanding the dynamics of RNA in Xenopus laevis oocytes. Fluorescence Recovery After Photobleaching (FRAP) is an experimental technique that is widely used to investigate the dynamics of molecular movement within cells by observing the recovery of fluorescence intensity in a photobleached region over time. To advance the understanding of RNA dynamics, we develop a reaction-diffusion-advection partial differential equation (PDE) model integrating both transport and diffusion mechanisms. We propose a pipeline for identifiability analysis to assess the model's ability to uniquely determine parameter values from observed FRAP data. Based on profile likelihood analysis and reparametrization, we examine the relationship between non- identifiable parameters, which improves the robustness of parameter estimation. We find out that the identifiability of the four parameters of interest is not exactly the same in different regions of the cell. Specifically, transport velocity and diffusion coefficient are identifiable in all regions of the cell, while some combinations of binding rate and unbinding rate are found to be identifiable near the nucleus.

math.AP↗

arXiv2025

Tactic: Adaptive Sparse Attention with Clustering and Distribution Fitting for Long-Context LLMs

Long-context models are essential for many applications but face inefficiencies in loading large KV caches during decoding. Prior methods enforce fixed token budgets for sparse attention, assuming a set number of tokens can approximate full attention. However, these methods overlook variations in the importance of attention across heads, layers, and contexts. To address these limitations, we propose Tactic, a sparsity-adaptive and calibration-free sparse attention mechanism that dynamically selects tokens based on their cumulative attention scores rather than a fixed token budget. By setting a target fraction of total attention scores, Tactic ensures that token selection naturally adapts to variations in attention sparsity. To efficiently approximate this selection, Tactic leverages clustering-based sorting and distribution fitting, allowing it to accurately estimate token importance with minimal computational overhead. We show that Tactic outperforms existing sparse attention algorithms, achieving superior accuracy and up to 7.29x decode attention speedup. This improvement translates to an overall 1.58x end-to-end inference speedup, making Tactic a practical and effective solution for long-context LLM inference in accuracy-sensitive applications.

cs.LG↗

arXiv2024

NanoFlow: Towards Optimal Large Language Model Serving Throughput

Large Language Models (LLMs) have resulted in a surging demand for planet-scale serving systems, where tens of thousands of GPUs continuously serve hundreds of millions of users. Consequently, throughput has emerged as a key metric that determines serving systems' performance. Due to large model sizes and memory-intensive self-attention, LLM serving has been commonly assumed to be memory-bound. Through a detailed analysis, we show that despite having memory-intensive components, end-to-end LLM serving is compute bound for most common workloads and LLMs. Alas, most existing serving engines fall short from optimal compute utilization, because the heterogeneous operations that comprise LLM serving--compute, memory, networking--are executed sequentially within a device. We propose NanoFlow, a novel serving framework that exploits intra-device parallelism, which overlaps the usage of heterogeneous resources within a single device. NanoFlow splits inputs into smaller nano-batches and duplicates operations to operate on each portion independently, enabling overlapping. NanoFlow automatically identifies the number, size, ordering, and GPU resource allocation of nano-batches to minimize the execution time, while considering the interference of concurrent operations. We evaluate NanoFlow's end-to-end serving throughput on several popular models such as LLaMA-2-70B, Mixtral 8x7B, LLaMA-3-8B, etc. With practical workloads, NanoFlow provides 1.91x throughput boost compared to state-of-the-art serving systems achieving 50% to 72% of optimal throughput across popular models.

cs.DC↗