arXiv subjects

Xiaoyi Qu

Publications and source records attributed to Xiaoyi Qu.

8 recordsLinked to original sources

arXiv2026

Low-Order Explicit Hessian Imitation Method for Large-Scale Supervised Machine Learning

An algorithm is proposed for solving optimization problems arising in neural network training for supervised learning. The unique feature of the algorithm is the use of an auxiliary loss, in addition to the original loss employed for model training. The purpose of the auxiliary loss is to provide a mechanism for creating a low-order Hessian-type approximation for the original loss. The proposed algorithm employs the resulting low-order second-derivative approximation terms in place of the second-order momentum terms (i.e., squared elements of the gradient of the loss function) in an overall scheme that has computational cost on par with an Adam-type approach. Whereas the squared elements of a gradient vector do not necessarily approximate second-order derivatives well, by careful construction of the auxiliary loss, second-order derivative-type approximations for the original loss can be computed and employed by the algorithm in an efficient manner. A convergence guarantee is provided for the proposed algorithm that is on par with guarantees available for similar stochastic diagonal-scaling methods. The results of numerical experiments show situations when the proposed algorithm outperforms Adam and other popular modern optimizers.

math.OC↗

arXiv2026

StableQAT: Stable Quantization-Aware Training at Ultra-Low Bitwidths

Quantization-aware training (QAT) is essential for deploying large models under strict memory and latency constraints, yet achieving stable and robust optimization at ultra-low bitwidths remains challenging. Common approaches based on the straight-through estimator (STE) or soft quantizers often suffer from gradient mismatch, instability, or high computational overhead. As such, we propose StableQAT, a unified and efficient QAT framework that stabilizes training in ultra low-bit settings via a novel, lightweight, and theoretically grounded surrogate for backpropagation derived from a discrete Fourier analysis of the rounding operator. StableQAT strictly generalizes STE as the latter arises as a special case of our more expressive surrogate family, yielding smooth, bounded, and inexpensive gradients that improve QAT training performance and stability across various hyperparameter choices. In experiments, StableQAT exhibits stable and efficient QAT at 2-4 bit regimes, demonstrating improved training stability, robustness, and superior performance with negligible training overhead against standard QAT techniques. Our code is available at https://github.com/microsoft/StableQAT.

cs.LG↗

arXiv2026

PRIME: Policy-Reinforced Iterative Multi-agent Execution for Algorithmic Reasoning in Large Language Models

Large language models have demonstrated remarkable capabilities across diverse reasoning tasks, yet their performance on algorithmic reasoning remains limited. To handle this limitation, we propose PRIME (Policy-Reinforced Iterative Multi-agent Execution), a framework comprising three specialized agents, an executor for step-by-step reasoning, a verifier for constraint checking, and a coordinator for backtracking control, optimized through group relative policy optimization. For comprehensive evaluation, we introduce PRIME-Bench, the largest algorithmic reasoning benchmark to date, comprising 86 tasks across 12 categories with 51,600 instances. Tasks span sorting algorithms, graph and tree structures, automata and state machines, symbolic reasoning, and constraint-based puzzles, with execution traces reaching over one million steps. Compared to baseline approach, PRIME improves average accuracy from 26.8% to 93.8%, a 250% relative gain. The largest improvements occur on tasks requiring sustained state tracking, with Turing machine simulation improving from 9% to 92% and long division from 16% to 94%. Ablation studies identify iterative verification as the primary contributor, preventing the error propagation that causes baseline approaches to fail catastrophically. Analysis across model scales (8B-120B parameters) reveals that smaller models benefit disproportionately, achieving accuracy comparable to models 8x larger.

cs.CL↗

arXiv2026

A Proximal-Gradient Method for Solving Regularized Optimization Problems with General Constraints

We propose, analyze, and test a proximal-gradient method for solving regularized optimization problems with general constraints. The method employs a decomposition strategy to compute trial steps and uses a merit function to determine step acceptance or rejection. Under various assumptions, we establish a worst-case iteration complexity result, prove that limit points are first-order KKT points, and show that manifold identification and active-set identification properties hold. Preliminary numerical experiments on a subset of the CUTEst test problems and sparse canonical correlation analysis problems demonstrate the promising performance of our approach.

math.OC↗

arXiv2025

Multi-Agent Medical Decision Consensus Matrix System: An Intelligent Collaborative Framework for Oncology MDT Consultations

Multidisciplinary team (MDT) consultations are the gold standard for cancer care decision-making, yet current practice lacks structured mechanisms for quantifying consensus and ensuring decision traceability. We introduce a Multi-Agent Medical Decision Consensus Matrix System that deploys seven specialized large language model agents, including an oncologist, a radiologist, a nurse, a psychologist, a patient advocate, a nutritionist and a rehabilitation therapist, to simulate realistic MDT workflows. The framework incorporates a mathematically grounded consensus matrix that uses Kendall's coefficient of concordance to objectively assess agreement. To further enhance treatment recommendation quality and consensus efficiency, the system integrates reinforcement learning methods, including Q-Learning, PPO and DQN. Evaluation across five medical benchmarks (MedQA, PubMedQA, DDXPlus, MedBullets and SymCat) shows substantial gains over existing approaches, achieving an average accuracy of 87.5% compared with 83.8% for the strongest baseline, a consensus achievement rate of 89.3% and a mean Kendall's W of 0.823. Expert reviewers rated the clinical appropriateness of system outputs at 8.9/10. The system guarantees full evidence traceability through mandatory citations of clinical guidelines and peer-reviewed literature, following GRADE principles. This work advances medical AI by providing structured consensus measurement, role-specialized multi-agent collaboration and evidence-based explainability to improve the quality and efficiency of clinical decision-making.

cs.MA↗

arXiv2025

HESSO: Towards Automatic Efficient and User Friendly Any Neural Network Training and Pruning

Structured pruning is one of the most popular approaches to effectively compress the heavy deep neural networks (DNNs) into compact sub-networks while retaining performance. The existing methods suffer from multi-stage procedures along with significant engineering efforts and human expertise. The Only-Train-Once (OTO) series has been recently proposed to resolve the many pain points by streamlining the workflow by automatically conducting (i) search space generation, (ii) structured sparse optimization, and (iii) sub-network construction. However, the built-in sparse optimizers in the OTO series, i.e., the Half-Space Projected Gradient (HSPG) family, have limitations that require hyper-parameter tuning and the implicit controls of the sparsity exploration, consequently requires intervening by human expertise. To address such limitations, we propose a Hybrid Efficient Structured Sparse Optimizer (HESSO). HESSO could automatically and efficiently train a DNN to produce a high-performing subnetwork. Meanwhile, it is almost tuning-free and enjoys user-friendly integration for generic training applications. To address another common issue of irreversible performance collapse observed in pruning DNNs, we further propose a Corrective Redundant Identification Cycle (CRIC) for reliably identifying indispensable structures. We numerically demonstrate the efficacy of HESSO and its enhanced version HESSO-CRIC on a variety of applications ranging from computer vision to natural language processing, including large language model. The numerical results showcase that HESSO can achieve competitive even superior performance to varying state-of-the-arts and support most DNN architectures. Meanwhile, CRIC can effectively prevent the irreversible performance collapse and further enhance the performance of HESSO on certain applications.

cs.LG↗

arXiv2025

Automatic Joint Structured Pruning and Quantization for Efficient Neural Network Training and Compression

Structured pruning and quantization are fundamental techniques used to reduce the size of deep neural networks (DNNs) and typically are applied independently. Applying these techniques jointly via co-optimization has the potential to produce smaller, high-quality models. However, existing joint schemes are not widely used because of (1) engineering difficulties (complicated multi-stage processes), (2) black-box optimization (extensive hyperparameter tuning to control the overall compression), and (3) insufficient architecture generalization. To address these limitations, we present the framework GETA, which automatically and efficiently performs joint structured pruning and quantization-aware training on any DNNs. GETA introduces three key innovations: (i) a quantization-aware dependency graph (QADG) that constructs a pruning search space for generic quantization-aware DNN, (ii) a partially projected stochastic gradient method that guarantees layerwise bit constraints are satisfied, and (iii) a new joint learning strategy that incorporates interpretable relationships between pruning and quantization. We present numerical experiments on both convolutional neural networks and transformer architectures that show that our approach achieves competitive (often superior) performance compared to existing joint pruning and quantization methods.

cs.LG↗

arXiv2024

A Proximal-Gradient Method for Constrained Optimization

We present a new algorithm for solving optimization problems with objective functions that are the sum of a smooth function and a (potentially) nonsmooth regularization function, and nonlinear equality constraints. The algorithm may be viewed as an extension of the well-known proximal-gradient method that is applicable when constraints are not present. To account for nonlinear equality constraints, we combine a decomposition procedure for computing trial steps with an exact merit function for determining trial step acceptance. Under common assumptions, we show that both the proximal parameter and merit function parameter eventually remain fixed, and then prove a worst-case complexity result for the maximum number of iterations before an iterate satisfying approximate first-order optimality conditions for a given tolerance is computed. Our preliminary numerical results indicate that our approach has great promise, especially in terms of returning approximate solutions that are structured (e.g., sparse solutions when a one-norm regularizer is used).

math.OC↗