arXiv subjects

Jingzhao Zhang

Publications and source records attributed to Jingzhao Zhang.

At least 37 records · Page 2Linked to original sources

arXiv2026

QuestA: Expanding Reasoning Capacity in LLMs via Question Augmentation

Reinforcement learning (RL) has emerged as a central paradigm for training large language models (LLMs) in reasoning tasks. Yet recent studies question RL's ability to incentivize reasoning capacity beyond the base model. This raises a key challenge: how can RL be adapted to solve harder reasoning problems more effectively? To address this challenge, we propose a simple yet effective strategy via Question Augmentation: introduce partial solutions during training to reduce problem difficulty and provide more informative learning signals. Our method, QuestA, when applied during RL training on math reasoning tasks, not only improves pass@1 but also pass@k-particularly on problems where standard RL struggles to make progress. This enables continual improvement over strong open-source models such as DeepScaleR and OpenMath Nemotron, further enhancing their reasoning capabilities. We achieve new state-of-the-art results on math benchmarks using 1.5B-parameter models: 72.50% (+10.73%) on AIME24, 62.29% (+12.79%) on AIME25, and 41.67% (+10.11%) on HMMT25. Code, data and model are available at https://github.com/foreverlasting1202/QuestA.

cs.CL↗

arXiv2026

Faster Gradient Methods for Highly-Smooth Stochastic Bilevel Optimization

This paper studies the complexity of finding an $ε$-stationary point for stochastic bilevel optimization when the upper-level problem is nonconvex and the lower-level problem is strongly convex. Recent work proposed the first-order method, F${}^2$SA, achieving the $\tilde{\mathcal{O}}(ε^{-6})$ upper complexity bound for first-order smooth problems. This is slower than the optimal $Ω(ε^{-4})$ complexity lower bound in its single-level counterpart. In this work, we show that faster rates are achievable for higher-order smooth problems. We first reformulate F$^2$SA as approximating the hyper-gradient with a forward difference. Based on this observation, we propose a class of methods F${}^2$SA-$p$ that uses $p$th-order finite difference for hyper-gradient approximation and improves the upper bound to $\tilde{\mathcal{O}}(p ε^{-4-p/2})$ for $p$th-order smooth problems. Finally, we demonstrate that the $Ω(ε^{-4})$ lower bound also holds for stochastic bilevel problems when the high-order smoothness holds for the lower-level variable, indicating that the upper bound of F${}^2$SA-$p$ is nearly optimal in the highly smooth region $p = Ω( \log ε^{-1} / \log \log ε^{-1})$.

math.OC↗

arXiv2025

Finite Sample Analyses for Continuous-time Linear Systems: System Identification and Online Control

Real world evolves in continuous time but computations are done from finite samples. Therefore, we study algorithms using finite observations in continuous-time linear dynamical systems. We first study the system identification problem, and propose a first non-asymptotic error analysis with finite observations. Our algorithm identifies system parameters without needing integrated observations over certain time intervals, making it more practical for real-world applications. Further we propose a lower bound result that shows our estimator is provably optimal up to constant factors. Moreover, we apply the above algorithm to online control regret analysis for continuous-time linear system. Our system identification method allows us to explore more efficiently, enabling the swift detection of ineffective policies. We achieve a regret of $\mathcal{O}(\sqrt{T})$ over a single $T$-time horizon in a controllable system, requiring only $\mathcal{O}(T)$ observations of the system.

eess.SY↗

arXiv2025

Multitask Battery Management with Flexible Pretraining

Industrial-scale battery management involves various types of tasks, such as estimation, prediction, and system-level diagnostics. Each task employs distinct data across temporal scales, sensor resolutions, and data channels. Building task-specific methods requires a great deal of data and engineering effort, which limits the scalability of intelligent battery management. Here we present the Flexible Masked Autoencoder (FMAE), a flexible pretraining framework that can learn with missing battery data channels and capture inter-correlations across data snippets. FMAE learns unified battery representations from heterogeneous data and can be adopted by different tasks with minimal data and engineering efforts. Experimentally, FMAE consistently outperforms all task-specific methods across five battery management tasks with eleven battery datasets. On remaining life prediction tasks, FMAE uses 50 times less inference data while maintaining state-of-the-art results. Moreover, when real-world data lack certain information, such as system voltage, FMAE can still be applied with marginal performance impact, achieving comparable results with the best hand-crafted features. FMAE demonstrates a practical route to a flexible, data-efficient model that simplifies real-world multi-task management of dynamical systems.

cs.LG↗

arXiv2025

Towards Black-Box Membership Inference Attack for Diffusion Models

Given the rising popularity of AI-generated art and the associated copyright concerns, identifying whether an artwork was used to train a diffusion model is an important research topic. The work approaches this problem from the membership inference attack (MIA) perspective. We first identify the limitation of applying existing MIA methods for proprietary diffusion models: the required access of internal U-nets. To address the above problem, we introduce a novel membership inference attack method that uses only the image-to-image variation API and operates without access to the model's internal U-net. Our method is based on the intuition that the model can more easily obtain an unbiased noise prediction estimate for images from the training set. By applying the API multiple times to the target image, averaging the outputs, and comparing the result to the original image, our approach can classify whether a sample was part of the training set. We validate our method using DDIM and Stable Diffusion setups and further extend both our approach and existing algorithms to the Diffusion Transformer architecture. Our experimental results consistently outperform previous methods.

cs.CR↗

arXiv2025

Understanding Nonlinear Implicit Bias via Region Counts in Input Space

One explanation for the strong generalization ability of neural networks is implicit bias. Yet, the definition and mechanism of implicit bias in non-linear contexts remains little understood. In this work, we propose to characterize implicit bias by the count of connected regions in the input space with the same predicted label. Compared with parameter-dependent metrics (e.g., norm or normalized margin), region count can be better adapted to nonlinear, overparameterized models, because it is determined by the function mapping and is invariant to reparametrization. Empirically, we found that small region counts align with geometrically simple decision boundaries and correlate well with good generalization performance. We also observe that good hyper-parameter choices such as larger learning rates and smaller batch sizes can induce small region counts. We further establish the theoretical connections and explain how larger learning rate can induce small region counts in neural networks.

cs.LG↗

arXiv2025

NeuralDB: Scaling Knowledge Editing in LLMs to 100,000 Facts with Neural KV Database

Efficiently editing knowledge stored in large language models (LLMs) enables model updates without large-scale training. One possible solution is Locate-and-Edit (L\&E), allowing simultaneous modifications of a massive number of facts. However, such editing may compromise the general abilities of LLMs and even result in forgetting edited facts when scaling up to thousands of edits. In this paper, we model existing linear L\&E methods as querying a Key-Value (KV) database. From this perspective, we then propose NeuralDB, an editing framework that explicitly represents the edited facts as a neural KV database equipped with a non-linear gated retrieval module, % In particular, our gated module only operates when inference involves the edited facts, effectively preserving the general abilities of LLMs. Comprehensive experiments involving the editing of 10,000 facts were conducted on the ZsRE and CounterFacts datasets, using GPT2-XL, GPT-J (6B) and Llama-3 (8B). The results demonstrate that NeuralDB not only excels in editing efficacy, generalization, specificity, fluency, and consistency, but also preserves overall performance across six representative text understanding and generation tasks. Further experiments indicate that NeuralDB maintains its effectiveness even when scaled to 100,000 facts (\textbf{50x} more than in prior work).

cs.CL↗

arXiv2025

Solving Convex-Concave Problems with $\tilde{\mathcal{O}}(ε^{-4/7})$ Second-Order Oracle Complexity

Previous algorithms can solve convex-concave minimax problems $\min_{x \in \mathcal{X}} \max_{y \in \mathcal{Y}} f(x,y)$ with $\mathcal{O}(ε^{-2/3})$ second-order oracle calls using Newton-type methods. This result has been speculated to be optimal because the upper bound is achieved by a natural generalization of the optimal first-order method. In this work, we show an improved upper bound of $\tilde{\mathcal{O}}(ε^{-4/7})$ by generalizing the optimal second-order method for convex optimization to solve the convex-concave minimax problem. We further apply a similar technique to lazy Hessian algorithms and show that our proposed algorithm can also be seen as a second-order ``Catalyst'' framework (Lin et al., JMLR 2018) that could accelerate any globally convergent algorithms for solving minimax problems.

math.OC↗

arXiv2025

Task Generalization With AutoRegressive Compositional Structure: Can Learning From $D$ Tasks Generalize to $D^{T}$ Tasks?

Large language models (LLMs) exhibit remarkable task generalization, solving tasks they were never explicitly trained on with only a few demonstrations. This raises a fundamental question: When can learning from a small set of tasks generalize to a large task family? In this paper, we investigate task generalization through the lens of autoregressive compositional structure, where each task is a composition of $T$ operations, and each operation is among a finite family of $D$ subtasks. This yields a total class of size $D^T$. We first show that generalization to all $D^T$ tasks is theoretically achievable by training on only $\widetilde{O}(D)$ tasks. Empirically, we demonstrate that Transformers achieve such exponential task generalization on sparse parity functions via In-context Learning (ICL) and chain-of-thought (CoT) reasoning. We further show generalization in arithmetic and translation, beyond parity functions.

cs.LG↗

arXiv2025

Online Control with Adversarial Disturbance for Continuous-time Linear Systems

We study online control for continuous-time linear systems with finite sampling rates, where the objective is to design an online procedure that learns under non-stochastic noise and performs comparably to a fixed optimal linear controller. We present a novel two-level online algorithm, by integrating a higher-level learning strategy and a lower-level feedback control strategy. This method offers a practical and robust solution for online control, which achieves sublinear regret. Our work provides the first nonasymptotic results for controlling continuous-time linear systems with finite number of interactions with the system. Moreover, we examine how to train an agent in domain randomization environments from a non-stochastic control perspective. By applying our method to the SAC (Soft Actor-Critic) algorithm, we achieved improved results in multiple reinforcement learning tasks within domain randomization environments. Our work provides new insights into non-asymptotic analyses of controlling continuous-time systems. Furthermore, our work brings practical intuition into controller learning under non-stochastic environments.

math.OC↗

arXiv2025

Fast and Multiphase Rates for Nearest Neighbor Classifiers

We study the scaling of classification error rates with respect to the size of the training dataset. In contrast to classical results where rates are minimax optimal for a problem class, this work starts with the empirical observation that, even for a fixed data distribution, the error scaling can have \emph{diverse} rates across different ranges of sample size. To understand when and why the error rate is non-uniform, we theoretically analyze nearest neighbor classifiers. We show that an error scaling law can have fine-grained rates: in the early phase, the test error depends polynomially on the data dimension and decreases fast; whereas in the later phase, the error depends exponentially on the data dimension and decreases slowly. Our analysis highlights the complexity of the data distribution in determining the test error. When the data are distributed benignly, we show that the generalization error of nearest neighbor classifier can depend polynomially, instead of exponentially, on the data dimension.

stat.ML↗

arXiv2025

Scalable Model Merging with Progressive Layer-wise Distillation

Model merging offers an effective way to integrate the capabilities of multiple fine-tuned models. However, the performance degradation of the merged model remains a challenge, particularly when none or few data are available. This paper first highlights the necessity of domain-specific data for model merging by proving that data-agnostic algorithms can have arbitrarily bad worst-case performance. Building on this theoretical insight, we explore the relationship between model merging and distillation, introducing a novel few-shot merging algorithm, ProDistill (Progressive Layer-wise Distillation). Unlike common belief that layer wise training hurts performance, we show that layer-wise teacher-student distillation not only enhances the scalability but also improves model merging performance. We conduct extensive experiments to show that compared to existing few-shot merging methods, ProDistill achieves state-of-the-art performance, with up to 6.14% and 6.61% improvements in vision and NLU tasks. Furthermore, we extend the experiments to models with over 10B parameters, showcasing the exceptional scalability of ProDistill.

cs.LG↗

arXiv2025

Second-Order Min-Max Optimization with Lazy Hessians

This paper studies second-order methods for convex-concave minimax optimization. Monteiro and Svaiter (2012) proposed a method to solve the problem with an optimal iteration complexity of $\mathcal{O}(ε^{-3/2})$ to find an $ε$-saddle point. However, it is unclear whether the computational complexity, $\mathcal{O}((N+ d^2) d ε^{-2/3})$, can be improved. In the above, we follow Doikov et al. (2023) and assume the complexity of obtaining a first-order oracle as $N$ and the complexity of obtaining a second-order oracle as $dN$. In this paper, we show that the computation cost can be reduced by reusing Hessian across iterations. Our methods take the overall computational complexity of $ \tilde{\mathcal{O}}( (N+d^2)(d+ d^{2/3}ε^{-2/3}))$, which improves those of previous methods by a factor of $d^{1/3}$. Furthermore, we generalize our method to strongly-convex-strongly-concave minimax problems and establish the complexity of $\tilde{\mathcal{O}}((N+d^2) (d + d^{2/3} κ^{2/3}) )$ when the condition number of the problem is $κ$, enjoying a similar speedup upon the state-of-the-art method. Numerical experiments on both real and synthetic datasets also verify the efficiency of our method.

math.OC↗

arXiv2025

From Sparse Dependence to Sparse Attention: Unveiling How Chain-of-Thought Enhances Transformer Sample Efficiency

Chain-of-thought (CoT) significantly enhances the reasoning performance of large language models (LLM). While current theoretical studies often attribute this improvement to increased expressiveness and computational capacity, we argue that expressiveness is not the primary limitation in the LLM regime, as current large models will fail on simple tasks. Using a parity-learning setup, we demonstrate that CoT can substantially improve sample efficiency even when the representation power is sufficient. Specifically, with CoT, a transformer can learn the function within polynomial samples, whereas without CoT, the required sample size is exponential. Additionally, we show that CoT simplifies the learning process by introducing sparse sequential dependencies among input tokens, and leads to a sparse and interpretable attention. We validate our theoretical analysis with both synthetic and real-world experiments, confirming that sparsity in attention layers is a key factor of the improvement induced by CoT.

cs.LG↗

arXiv2025

Functionally Constrained Algorithm Solves Convex Simple Bilevel Problems

This paper studies simple bilevel problems, where a convex upper-level function is minimized over the optimal solutions of a convex lower-level problem. We first show the fundamental difficulty of simple bilevel problems, that the approximate optimal value of such problems is not obtainable by first-order zero-respecting algorithms. Then we follow recent works to pursue the weak approximate solutions. For this goal, we propose a novel method by reformulating them into functionally constrained problems. Our method achieves near-optimal rates for both smooth and nonsmooth problems. To the best of our knowledge, this is the first near-optimal algorithm that works under standard assumptions of smoothness or Lipschitz continuity for the objective functions.

math.OC↗

arXiv2024

On the Overlooked Pitfalls of Weight Decay and How to Mitigate Them: A Gradient-Norm Perspective

Weight decay is a simple yet powerful regularization technique that has been very widely used in training of deep neural networks (DNNs). While weight decay has attracted much attention, previous studies fail to discover some overlooked pitfalls on large gradient norms resulted by weight decay. In this paper, we discover that, weight decay can unfortunately lead to large gradient norms at the final phase (or the terminated solution) of training, which often indicates bad convergence and poor generalization. To mitigate the gradient-norm-centered pitfalls, we present the first practical scheduler for weight decay, called the Scheduled Weight Decay (SWD) method that can dynamically adjust the weight decay strength according to the gradient norm and significantly penalize large gradient norms during training. Our experiments also support that SWD indeed mitigates large gradient norms and often significantly outperforms the conventional constant weight decay strategy for Adaptive Moment Estimation (Adam).

cs.LG↗

arXiv2024

Random Masking Finds Winning Tickets for Parameter Efficient Fine-tuning

Fine-tuning large language models (LLM) can be costly. Parameter-efficient fine-tuning (PEFT) addresses the problems by training a fraction of the parameters, whose success reveals the expressiveness and flexibility of pretrained models. This paper studies the limit of PEFT, by further simplifying its design and reducing the number of trainable parameters beyond standard setups. To this end, we use Random Masking to fine-tune the pretrained model. Despite its simplicity, we show that Random Masking is surprisingly effective: with a larger-than-expected learning rate, Random Masking can match the performance of standard PEFT algorithms such as LoRA on various tasks, using fewer trainable parameters. We provide both empirical and theoretical explorations into the success of Random Masking. We show that masking induces a flatter loss landscape and more distant solutions, which allows for and necessitates large learning rates.

cs.LG↗

arXiv2024

A Quadratic Synchronization Rule for Distributed Deep Learning

In distributed deep learning with data parallelism, synchronizing gradients at each training step can cause a huge communication overhead, especially when many nodes work together to train large models. Local gradient methods, such as Local SGD, address this issue by allowing workers to compute locally for $H$ steps without synchronizing with others, hence reducing communication frequency. While $H$ has been viewed as a hyperparameter to trade optimization efficiency for communication cost, recent research indicates that setting a proper $H$ value can lead to generalization improvement. Yet, selecting a proper $H$ is elusive. This work proposes a theory-grounded method for determining $H$, named the Quadratic Synchronization Rule (QSR), which recommends dynamically setting $H$ in proportion to $\frac{1}{η^2}$ as the learning rate $η$ decays over time. Extensive ImageNet experiments on ResNet and ViT show that local gradient methods with QSR consistently improve the test accuracy over other synchronization strategies. Compared with the standard data parallel training, QSR enables Local AdamW on ViT-B to cut the training time on 16 or 64 GPUs down from 26.7 to 20.2 hours or from 8.6 to 5.5 hours and, at the same time, achieves $1.16\%$ or $0.84\%$ higher top-1 validation accuracy.

cs.LG↗