arXiv subjects

Xiaobing Feng

Publications and source records attributed to Xiaobing Feng.

At least 19 recordsLinked to original sources

arXiv2026

Unlocking Multi-Component Bulk-Materials Molecular Dynamics with a Small-Footprint Machine Learning Interatomic Potential

Bulk materials, as opposed to nanomaterials, require molecular dynamics (MD) simulations on a large spatial scale (~10^9 atoms or more) to adequately capture their atomic-scale physical properties. Previously, the introduction of machine-learning interatomic potentials (MLIPs) has extended MD to this scale, but even single-component bulk systems require tens of thousands of GPUs on high-end supercomputers. However, multi-component bulk MD simulations remain barely achievable, as the HBM footprint of existing MLIPs - already substantial for single-component systems - grows explosively in multi-component scenarios. This paper proposes an MLIP with a small HBM footprint - less than 3% that of existing MLIPs - unlocking multi-component bulk MD using only hundreds of GPUs. This is achieved by first identifying feature vectors and intermediate tensors as the two primary contributors to HBM footprints in existing MLIPs. To address these two sources, the dimensionality of the feature vectors has been reduced by introducing physical and chemical knowledge, and intermediate tensors have been eliminated by aggressively fusing all kernels into a single mega-kernel. In evaluation, the proposed MLIP has used 144 NVIDIA A100 GPUs to perform MD simulations on a 6-component bulk system with 1.14x10^9 atoms, while previously such MD simulation spatial scale has been restricted to unary systems and typically achieved on high-end supercomputers equipped with tens of thousands of GPUs.

physics.comp-ph↗

arXiv2026

Decode-Time Grammars: Constrained LLM Generation over a Refinement Order of Grammar Fragments

Large language models now write a growing share of the world's code, increasingly inside agents and serving systems that compile, execute, or dispatch generated code without line-by-line review. This works well for mainstream languages but remains brittle for low-resource programming surfaces such as domain-specific languages, custom library APIs, and command-line tools. Even under grammar-constrained decoding, a model can still produce references invalid in the current environment: a buffer never declared, a column absent from the schema, a function the library does not provide, or an unsupported CLI option. This paper introduces decode-time grammars: grammar fragments instantiated during generation from a runtime environment Gamma. A region-specific policy selects a fragment for each hole, and a tightening operator replaces open reference positions with Gamma-typed slots whose candidates are exactly the names, fields, APIs, or options available at that point. Newly generated declarations enter Gamma before later regions are decoded, so the constraining grammar can depend on the prefix already generated. This ensures not only grammatical correctness but also semantic correctness, by preventing references to undefined symbols. We formalize grammar fragments as environment-indexed grammars ordered by refinement, prove No-Ghost soundness for Gamma-slotted fragments, show that refinement preserves this support-set guarantee, and characterize the boundary of mask-enforceable properties. We implement the approach in gproj with offline grammar induction and online policy resolution. Across TileLang, SQL, and P4, with models from 0.6B to 236B parameters, gproj eliminates ghost references by construction at moderate overhead over standard constrained decoding.

cs.PL↗

arXiv2026

Learning When to Optimize: Verified Optimization Skills from Expert GPU-Kernel Lineages

LLM-based agents are increasingly used to generate GPU kernels, but they often know what optimizations to try without knowing when those optimizations are sound. We introduce KLineage, which learns this missing "when" knowledge from expert kernels: instead of relying on forward rollouts, KLineage walks expert implementations backward through validation-gated simplifications and reverses each accepted step into a reusable optimization skill. Each skill records not only the optimization intent, but also where it applies in code, what conditions made it valid, what effect it had, and what failures its assumptions avoid. A downstream LLM materializes these skills on new code surfaces under the same compile/correctness/profile gate. On five expert workloads across two NVIDIA architectures, these lineage-derived skills serve as an effective optimization curriculum, exceeding recent memory-based LLM-kernel baselines in both final kernel quality and optimization efficiency under the same fixed budget. We additionally use a separate 22-instance held-out check as a sanity test against source-case memorization.

cs.AI↗

arXiv2026

Beyond Pass-by-Pass Optimization: Intent-Driven IR Optimization with Large Language Models

Modern compilers optimize programs through a sequence of modular passes over intermediate representations (IR). While this pass-by-pass paradigm offers engineering benefits, it suffers from a pass coordination problem: locally beneficial transformations may block more profitable optimizations in later stages. This limitation stems from the lack of an explicit notion of optimization intent, defined as a holistic strategy for coordinating multiple transformations toward a global performance objective. Recent LLM-based approaches formulate IR optimization as an end-to-end generation task, thereby avoiding the traditional pass-by-pass structure. However, optimization intent remains implicit in these methods, forcing models to jointly infer optimization strategy and generate low-level transformations, which limits both correctness and performance. We propose IntOpt, the first intent-driven IR optimizer that explicitly separates high-level optimization intent from low-level analysis and transformation. IntOpt organizes IR optimization into three stages: intent formulation, intent refinement, and intent realization, enabling globally coordinated transformations. Experiments show that IntOpt achieves 90.5% verified correctness and 2.660x average speedup on 200-program test set, outperforming state-of-the-art LLM-based optimizers in both correctness and performance, and surpassing modern compiler with the -O3 option on 37 benchmarks with speedups of up to 272.60x.

cs.PL↗

arXiv2026

The New Compiler Stack: A Survey on the Synergy of LLMs and Compilers

This survey has provided a systematic overview of the emerging field of LLM-enabled compilation by addressing several key research questions. We first answered how LLMs are being integrated by proposing a comprehensive, multi-dimensional taxonomy that categorizes works based on their Design Philosophy (Selector, Translator, Generator), LLM Methodology, their operational Level of Code Abstraction, and the specific Task Type they address. In answering what advancements these approaches offer, we identified three primary benefits: the democratization of compiler development, the discovery of novel optimization strategies, and the broadening of the compiler's traditional scope. Finally, in addressing the field's challenges and opportunities, we highlighted the critical hurdles of ensuring correctness and achieving scalability, while identifying the development of hybrid systems as the most promising path forward. By providing these answers, this survey serves as a foundational roadmap for researchers and practitioners, charting the course for a new generation of LLM-powered, intelligent, adaptive and synergistic compilation tools.

cs.PL↗

arXiv2025

Full moment error estimates in strong norms for numerical approximations of stochastic Navier-Stokes equations with multiplicative noise, Part I: time discretization

This paper focuses on deriving optimal-order full moment error estimates in strong norms for both velocity and pressure approximations in the Euler-Maruyama time discretization of the stochastic Navier-Stokes equations with multiplicative noise. Additionally, it introduces a novel approach and framework for the numerical analysis of nonlinear stochastic partial differential equations (SPDEs) with multiplicative noise in general. The main ideas of this approach include establishing exponential stability estimates for the SPDE solution, leveraging a discrete stochastic Gronwall inequality, and employing a bootstrap argument.

math.NA↗

arXiv2025

LEGO-Compiler: Enhancing Neural Compilation Through Translation Composability

Large language models (LLMs) have the potential to revolutionize how we design and implement compilers and code translation tools. However, existing LLMs struggle to handle long and complex programs. We introduce LEGO-Compiler, a novel neural compilation system that leverages LLMs to translate high-level languages into assembly code. Our approach centers on three key innovations: LEGO translation, which decomposes the input program into manageable blocks; breaking down the complex compilation process into smaller, simpler verifiable steps by organizing it as a verifiable LLM workflow by external tests; and a feedback mechanism for self-correction. Supported by formal proofs of translation composability, LEGO-Compiler demonstrates high accuracy on multiple datasets, including over 99% on ExeBench and 97.9% on industrial-grade AnsiBench. Additionally, LEGO-Compiler has also acheived near one order-of-magnitude improvement on compilable code size scalability. This work opens new avenues for applying LLMs to system-level tasks, complementing traditional compiler technologies.

cs.PL↗

arXiv2025

Optimal Order Space-Time Discretization Methods for the Nonlinear Stochastic Elastic Wave Equations with Multiplicative Noise

This paper develops and analyzes an optimal-order semi-discrete scheme and its fully discrete finite element approximation for nonlinear stochastic elastic wave equations with multiplicative noise. A non-standard time-stepping scheme is introduced for time discretization, it is showed that the scheme converges with rates $O(τ)$ and $O(τ^{\frac32})$ respectively in the energy- and $L^2$-norm, which are optimal with respect to the time regularity of the PDE solution. For spatial discretization, the standard finite element method is employed. It is proven that the fully discrete method converges with optimal rates $O(τ+ h)$ and $O(τ^{\frac{3}{2}} + h^2)$ respectively in the energy- and $L^2$-norm. The cruxes of the analysis are to establish some high-moment stability results and utilize a refined error estimate for the trapezoidal quadrature rule to control the nonlinearities from the drift term and the multiplicative noise. Numerical experiments are also provided to validate the theoretical results.

math.NA↗

arXiv2025

When Grammar Guides the Attack: Uncovering Control-Plane Vulnerabilities in LLMs with Structured Output

Content Warning: This paper may contain unsafe or harmful content generated by LLMs that may be offensive to readers. Large Language Models (LLMs) increasingly serve as tooling platforms through structured output APIs, but the grammar-guided decoding that powers this feature opens a critical control-plane attack surface orthogonal to traditional data-plane vulnerabilities. We introduce Constrained Decoding Attack (CDA), a new jailbreak class that targets the LLM control plane. CDA is best characterized as a control-to-semantic pipeline: (1) schema-enforced logit masking injects a malicious prefix into the generation trajectory, and (2) the model itself completes the harmful intent. Unlike data-plane jailbreaks that rely on bypassing alignment with visible inputs, CDA acts on the decoding process itself, so internal safety alignment alone cannot stop it. We instantiate CDA with EnumAttack, which hides malicious content in enum fields, and the more evasive DictAttack, which decouples the payload across a benign prompt and a dictionary-based grammar. Across 13 proprietary/open-weight models and five standard benchmarks, DictAttack achieves 94.3--99.5% Attack Success Rate (ASR) on flagship models including gpt-5, gemini-2.5-pro, deepseek-r1, and gpt-oss-120b. While basic grammar auditing mitigates EnumAttack, DictAttack still sustains 75.8% ASR against SOTA jailbreak guardrails, exposing a "semantic gap" that demands cross-plane defenses bridging the data and control planes. Project page and code are available at https://ict-cda.github.io/.

cs.CR↗

arXiv2024

$\Gamma$-convergence of an Enhanced Finite Element Method for Mani\`a's and Foss's Problems Exhibiting the Lavrentiev Gap Phenomenon

It is well-known that numerically approximating calculus of variations problems possessing a Lavrentiev Gap Phenomenon (LGP) is challenging, and the standard numerical methodologies, such as finite element, finite difference, and discontinuous Galerkin methods, fail to give convergent methods because they cannot overcome the gap. This paper is a continuation of a 2016 paper by Feng and Schnake, where a promising enhanced finite element method was proposed to overcome the LGP in the classical Mani\`a's problem. The first goal of this paper is to provide a complete $\Gamma$-convergence proof for this enhanced finite element method, hence, establishing a theoretical foundation for the method. The crux of the convergence analysis is taking advantage of the regularity of the minimizer and viewing the minimization problem as posed over the fractional Sobolev space $W^{1 + s, p}(0, 1)$ (for $s > 0$) rather than the original admissible space $W^{1, p}(0, 1)$. The second goal is to extend the enhanced finite element method to the two-dimensional Foss's problem from nonlinear elasticity, which is also known to possess the LGP, and to establish its $\Gamma$-convergence as well.

math.NA↗

arXiv2024

Optimal order time discretizations for stochastic semilinear wave equations with multiplicative noise

This paper is concerned with developing and analyzing two novel implicit temporal discretization methods for the stochastic semilinear wave equations with multiplicative noise. The proposed methods are natural extensions of well-known time-discrete schemes for deterministic wave equations, hence, they are easy to implement. It is proved that both methods are energy-stable. Moreover, the first method is shown to converge with the linear order in the energy norm, while the second method converges with the $\mathcal{O}(τ^{\frac32})$ order in the $L^2$-norm, which is optimal with respect to the time regularity of the solution to the underlying stochastic PDE. The convergence analyses of both methods, which are different and quite involved, require some novel numerical techniques to overcome difficulties caused by the nonlinear noise term and the interplay between nonlinear drift and diffusion. Numerical experiments are provided to validate the sharpness of the theoretical error estimate results.

math.NA↗

arXiv2024

M4: Multi-Proxy Multi-Gate Mixture of Experts Network for Multiple Instance Learning in Histopathology Image Analysis

Multiple instance learning (MIL) has been successfully applied for whole slide images (WSIs) analysis in computational pathology, enabling a wide range of prediction tasks from tumor subtyping to inferring genetic mutations and multi-omics biomarkers. However, existing MIL methods predominantly focus on single-task learning, resulting in not only overall low efficiency but also the overlook of inter-task relatedness. To address these issues, we proposed an adapted architecture of Multi-gate Mixture-of-experts with Multi-proxy for Multiple instance learning (M4), and applied this framework for simultaneous prediction of multiple genetic mutations from WSIs. The proposed M4 model has two main innovations: (1) utilizing a mixture of experts with multiple gating strategies for multi-genetic mutation prediction on a single pathological slide; (2) constructing multi-proxy expert network and gate network for comprehensive and effective modeling of pathological image information. Our model achieved significant improvements across five tested TCGA datasets in comparison to current state-of-the-art single-task methods. The code is available at:https://github.com/Bigyehahaha/M4.

cs.CV↗

arXiv2024

A Tale of Two Paths: Toward a Hybrid Data Plane for Efficient Far-Memory Applications

With rapid advances in network hardware, far memory has gained a great deal of traction due to its ability to break the memory capacity wall. Existing far memory systems fall into one of two data paths: one that uses the kernel's paging system to transparently access far memory at the page granularity, and a second that bypasses the kernel, fetching data at the object granularity. While it is generally believed that object fetching outperforms paging due to its fine-grained access, it requires significantly more compute resources to run object-level LRU and eviction. We built Atlas, a hybrid data plane enabled by a runtime-kernel co-design that simultaneously enables accesses via these two data paths to provide high efficiency for real-world applications. Atlas uses always-on profiling to continuously measure page locality. For workloads already with good locality, paging is used to fetch data, whereas for those without, object fetching is employed. Object fetching moves objects that are accessed close in time to contiguous local space, dynamically improving locality and making the execution increasingly amenable to paging, which is much more resource-efficient. Our evaluation shows that Atlas improves the throughput (e.g., by 1.5x and 3.2x) and reduces the tail latency (e.g., by one and two orders of magnitude) when using remote memory, compared with AIFM and Fastswap, the state-of-the-art techniques respectively in the two categories.

cs.DC↗

arXiv2024

Mathematical and numerical analysis for PDE systems modeling intravascular drug release from arterial stents and transport in arterial tissue

This paper is concerned with the PDE and numerical analysis of a modified one-dimensional intravascular stent model originally proposed in [4]. It is proved that the modified model has a unique weak solution using the Galerkin method combined with a compactness argument. A semi-discrete finite element method and a fully discrete scheme using the Euler time-stepping are formulated for the PDE model. Optimal order error estimates in the energy norm are proved for both schemes. Numerical results are presented along with comparisons between different decoupling strategies and time-stepping schemes. Lastly, extensions of the model and its PDE and numerical analysis results to the two-dimensional case are also briefly discussed.

math.NA↗

arXiv2024

An efficient implementation algorithm for quasi-Monte Carlo approximations of high-dimensional integrals

In this paper, we develop and test a fast numerical algorithm, called MDI-LR, for efficient implementation of quasi-Monte Carlo lattice rules for computing $d$-dimensional integrals of a given function. It is based on the idea of converting and improving the underlying lattice rule into a tensor product rule by an affine transformation and adopting the multilevel dimension iteration approach which computes the function evaluations (at the integration points) in the tensor product multi-summation in cluster and iterates along each (transformed) coordinate direction so that a lot of computations can be reused. The proposed algorithm also eliminates the need for storing integration points and computing function values independently at each point. Extensive numerical experiments are presented to gauge the performance of the algorithm MDI-LR and to compare it with standard implementation of quasi-Monte Carlo lattice rules. It is also showed numerically that the algorithm MDI-LR can achieve a computational complexity of order $O(N^2d^3)$ or better, where $N$ represents the number of points in each (transformed) coordinate direction and $d$ standard for the dimension. Thus, the algorithm MDI-LR effectively overcomes the curse of dimensionality and revitalizes QMC lattice rules for high-dimensional integration.

math.NA↗

arXiv2022

An efficient and fast sparse grid algorithm for high-dimensional numerical integration

This paper is concerned with developing an efficient numerical algorithm for fast implementation of the sparse grid method for computing the $d$-dimensional integral of a given function. The new algorithm, called the MDI-SG ({\em multilevel dimension iteration sparse grid}) method, implements the sparse grid method based on a dimension iteration/reduction procedure, it does not need to store the integration points, neither does it compute the function values independently at each integration point, instead, it re-uses the computation for function evaluations as much as possible by performing the function evaluations at all integration points in a cluster and iteratively along coordinate directions. It is shown numerically that the computational complexity (in terms of CPU time) of the proposed MDI-SG method is of polynomial order $O(Nd^3 )$ or better, compared to the exponential order $O(N(\log N)^{d-1})$ for the standard sparse grid method, where $N$ denotes the maximum number of integration points in each coordinate direction. As a result, the proposed MDI-SG method effectively circumvents the curse of dimensionality suffered by the standard sparse grid method for high-dimensional numerical integration.

math.NA↗

arXiv2022

A fast multilevel dimension iteration algorithm for high dimensional numerical integration

In this paper, we propose and study a fast multilevel dimension iteration (MDI) algorithm for computing arbitrary $d$-dimensional integrals based on tensor product approximations. It reduces the computational complexity (in terms of the CPU time) of a tensor product method from the exponential order $O(N^d)$ to the polynomial order {\color{black} $O(d^3N^2)$ or better}, where $N$ stands for the number of quadrature points in each coordinate direction. As a result, the proposed MDI algorithm effectively circumvents the curse of the dimensionality of tensor product methods for high dimensional numerical integration. The main idea of the proposed MDI algorithm is to compute the function evaluations at all integration points in the cluster and iteratively along each coordinate direction, so lots of computations for function evaluations can be reused in each iteration. This idea is also applicable to any quadrature rule whose integration points have a lattice-like structure.

math.NA↗

arXiv2022

High moment and pathwise error estimates for fully discrete mixed finite element approximattions of stochastic Navier-Stokes equations with additive noise

This paper is concerned with high moment and pathwise error estimates for fully discrete mixed finite element approximattions of stochastic Navier-Stokes equations with general additive noise. The implicit Euler-Maruyama scheme and standard mixed finite element methods are employed respectively for the time and space discretizations. High moment error estimates for both velocity and a time-avraged pressure approximations in strong $L^2$ and energy norms are obtained, pathwise error estimates are derived by using the Kolmogorov Theorem. Unlike their derterministic counterparts, the spatial error constants grow in the order of $O(k^{-\frac12})$, where $k$ denotes time step size. Numerical experiments are also provided to validate the error estimates and their sharpness.

math.NA↗