arXiv subjects

Shuolin Li

Publications and source records attributed to Shuolin Li.

9 recordsLinked to original sources

arXiv2026

Revisiting RaBitQ and TurboQuant: A Symmetric Comparison of Methods, Theory, and Experiments

This technical note revisits the relationship between RaBitQ and TurboQuant under a unified comparison framework. We compare the two methods in terms of methodology, theoretical guarantees, and empirical performance, using a reproducible, transparent, and symmetric setup. Our results show that, despite the claimed advantage of TurboQuant, TurboQuant performs worse than RaBitQ in most tested settings of inner-product estimation, nearest-neighbor search and KV cache quantization. We further find that several reported runtime and recall results in the TurboQuant paper could not be reproduced from the released implementation under the stated configuration. Overall, this note clarifies the shared structure and genuine differences between the two lines of work, while documenting reproducibility issues in the experimental results reported by the TurboQuant paper.

cs.LG↗

arXiv2025

Calibrating Geophysical Predictions under Constrained Probabilistic Distributions

Machine learning (ML) has shown significant promise in studying complex geophysical dynamical systems, including turbulence and climate processes. Such systems often display sensitive dependence on initial conditions, reflected in positive Lyapunov exponents, where even small perturbations in short-term forecasts can lead to large deviations in long-term outcomes. Thus, meaningful inference requires not only accurate short-term predictions, but also consistency with the system's long-term attractor that is captured by the marginal distribution of state variables. Existing approaches attempt to address this challenge by incorporating spatial and temporal dependence, but these strategies become impractical when data are extremely sparse. In this work, we show that prior knowledge of marginal distributions offers valuable complementary information to short-term observations, motivating a distribution-informed learning framework. We introduce a calibration algorithm based on normalization and the Kernelized Stein Discrepancy (KSD) to enhance ML predictions. The method here employs KSD within a reproducing kernel Hilbert space to calibrate model outputs, improving their fidelity to known physical distributions. This not only sharpens pointwise predictions but also enforces consistency with non-local statistical structures rooted in physical principles. Through synthetic experiments-spanning offline climatological CO2 fluxes and online quasi-geostrophic flow simulations-we demonstrate the robustness and broad utility of the proposed framework.

physics.ao-ph↗

arXiv2025

Machine Learning Workflows in Climate Modeling: Design Patterns and Insights from Case Studies

Machine learning has been increasingly applied in climate modeling on system emulation acceleration, data-driven parameter inference, forecasting, and knowledge discovery, addressing challenges such as physical consistency, multi-scale coupling, data sparsity, robust generalization, and integration with scientific workflows. This paper analyzes a series of case studies from applied machine learning research in climate modeling, with a focus on design choices and workflow structure. Rather than reviewing technical details, we aim to synthesize workflow design patterns across diverse projects in ML-enabled climate modeling: from surrogate modeling, ML parameterization, probabilistic programming, to simulation-based inference, and physics-informed transfer learning. We unpack how these workflows are grounded in physical knowledge, informed by simulation data, and designed to integrate observations. We aim to offer a framework for ensuring rigor in scientific machine learning through more transparent model development, critical evaluation, informed adaptation, and reproducibility, and to contribute to lowering the barrier for interdisciplinary collaboration at the interface of data science and climate modeling.

cs.LG↗

arXiv2025

Integer Linear Programming Preprocessing for Maximum Satisfiability

The Maximum Satisfiability problem (MaxSAT) is a major optimization challenge with numerous practical applications. In recent MaxSAT evaluations, most MaxSAT solvers have incorporated an Integer Linear Programming (ILP) solver into their portfolios. However, a good portfolio strategy requires a lot of tuning work and is limited to the profiling benchmark. This paper proposes a methodology to fully integrate ILP preprocessing techniques into the MaxSAT solving pipeline and investigates the impact on the top-performing MaxSAT solvers. Experimental results show that our approach helps to improve 5 out of 6 state-of-the-art MaxSAT solvers, especially for WMaxCDCL-OpenWbo1200, the winner of the MaxSAT evaluation 2024 on the unweighted track, which is able to solve 15 additional instances using our methodology.

cs.AI↗

arXiv2024

Deep Generative Data Assimilation in Multimodal Setting

Robust integration of physical knowledge and data is key to improve computational simulations, such as Earth system models. Data assimilation is crucial for achieving this goal because it provides a systematic framework to calibrate model outputs with observations, which can include remote sensing imagery and ground station measurements, with uncertainty quantification. Conventional methods, including Kalman filters and variational approaches, inherently rely on simplifying linear and Gaussian assumptions, and can be computationally expensive. Nevertheless, with the rapid adoption of data-driven methods in many areas of computational sciences, we see the potential of emulating traditional data assimilation with deep learning, especially generative models. In particular, the diffusion-based probabilistic framework has large overlaps with data assimilation principles: both allows for conditional generation of samples with a Bayesian inverse framework. These models have shown remarkable success in text-conditioned image generation or image-controlled video synthesis. Likewise, one can frame data assimilation as observation-conditioned state calibration. In this work, we propose SLAMS: Score-based Latent Assimilation in Multimodal Setting. Specifically, we assimilate in-situ weather station data and ex-situ satellite imagery to calibrate the vertical temperature profiles, globally. Through extensive ablation, we demonstrate that SLAMS is robust even in low-resolution, noisy, and sparse data settings. To our knowledge, our work is the first to apply deep generative framework for multimodal data assimilation using real-world datasets; an important step for building robust computational simulators, including the next-generation Earth system models. Our code is available at: https://github.com/yongquan-qu/SLAMS

cs.CV↗

arXiv2021

Branching Strategy Selection Approach Based on Vivification Ratio

The two most effective branching strategies LRB and VSIDS perform differently on different types of instances. Generally, LRB is more effective on crafted instances, while VSIDS is more effective on application ones. However, distinguishing the types of instances is difficult. To overcome this drawback, we propose a branching strategy selection approach based on the vivification ratio. This approach uses the LRB branching strategy more to solve the instances with a very low vivification ratio. We tested the instances from the main track of SAT competitions in recent years. The results show that the proposed approach is robust and it significantly increases the number of solved instances. It is worth mentioning that, with the help of our approach, the solver Maple\_CM can solve more than 16 instances for the benchmark from the 2020 SAT competition.

cs.AI↗

arXiv2021

Roughness-induced critical phenomenon analogy for turbulent friction factor explained by a co-spectral budget model

Drawing on an analogy to critical phenomena, it was shown that the Nikuradse turbulent friction factor ($f_t$) measurements in pipes of radius $R$ and wall roughness $r$ can be collapsed onto a one-dimensional curve expressed as a conveyance law $f_t Re^{1/4}=g_o(χ)$, where $Re$ is a bulk Reynolds number, $χ=Re^{3/4}\left({r}/{R}\right)$. The implicit function $g_o(.)$ was conjectured based on matching two asymptotic limits of $f_t$. However, the connection between $g_o(.)$ and the phenomenon it proclaims to represent - turbulent eddies - remains lacking. Using models for the wall-normal velocity spectrum and return-to-isotropy for pressure-strain effects to close a co-spectral density budget, a derivation of $g_o(.)$ is offered. The proposed method explicitly derives the solution of the conveyance law and provides a physical interpretation of $χ$ as a dimensionless length scale reflecting the competition between viscous sublayer thickness and characteristic height of roughness elements. The application of the proposed method to other published measurements spanning roughness and Reynolds numbers beyond the original Nikuradse range is further discussed.

physics.flu-dyn↗

arXiv2021

The Role of Geographic Spreaders in Infectious Pattern Formation and Front Propagation Speeds

The pattern formation and spatial spread of infectious populations are investigated using a kernel-based Susceptible-Infectious-Recovered (SIR) model applicable across a wide range of basic reproduction numbers $R_o$. The focus is on the role of geographic spreaders defined here as a portion of the infected population ($ϕ$) experiencing high mobility between identical communities. The spatial organization of the infected population and invasive front speeds ($c_{max}$) are determined when the infections are randomly initiated in space within multiple communities. For small but finite $ϕ$, scaling analysis in 1-dimension and simulation results in 2-dimensions suggest that $c_{max}\sim (1-ϕ) γ(R_o-1) σ$, where $γ$ is the inverse of the infectious duration, and $σ^2$ is the variance of the spatial kernel describing mobility of long-distance spreaders across communities. Hence, $c_{max}$ is not significantly affected by the small $ϕ$ though reductions in $ϕ$ act as retardation factors to the attainment of $c_{max}$. The $σ$ determines the spatial organization of infections across communities. When $σ>5dr$ (long-distance mobility, where $dr$ is the minimum spatial extent defining adjacent communities), the infectious population will experience a transient but spatially coherent pattern with a wavelength that can be derived from the spreading kernel properties.

physics.soc-ph↗

arXiv2021

A co-spectral budget model links turbulent eddies to suspended sediment concentration in channel flows

The vertical distribution of suspended sediment concentration (SSC) remains a subject of active research given its relevance to a plethora of problems in hydraulics, hydrology, ecology, and water quality control. Much of the classical theories developed over the course of 90 years represent the effects of turbulence on suspended sediments (SS) using an effective mixing length or eddy diffusivity without explicitly accounting for the energetics of turbulent eddies across scales. To address this gap, the turbulent flux of sediments is derived using a co-spectral budget (CSB) model that can be imminently used in SS and other fine particle transport models. The CSB closes the pressure-redistribution effect using a spectral linear Rotta scheme modified to include isotropoziation of production and interactions between turbulent eddies and sediment grains through a modified scale-dependent de-correlation time. The result is a formulation similar in complexity to the widely used Rouse's equation but with all characteristic scales, Reynolds number, and Schmidt number effects derived from well-established spectral shapes of the vertical velocity and accepted constants from turbulence models. Finally, the proposed CSB model can recover Prandtl's and Rouse's equations under restricted conditions.

physics.flu-dyn↗