arXiv subjects

Chu Wang

Publications and source records attributed to Chu Wang.

At least 37 records · Page 2Linked to original sources

arXiv2026

Real-Time Trend Prediction via Continually-Aligned LLM Query Generation

Trending news detection in low-traffic search environments faces a fundamental cold-start problem, where a lack of query volume prevents systems from identifying emerging or long-tail trends. Existing methods relying on keyword frequency or query spikes are inherently slow and ineffective in these sparse settings, lagging behind real-world shifts in attention. We introduce RTTP, a novel Real-Time Trending Prediction framework that generates search queries directly from news content instead of waiting for users to issue them. RTTP leverages a continual learning LLM (CL-LLM) that converts posts into search-style queries and scores them using engagement strength + creator authority, enabling early trend surfacing before search volume forms. To ensure adaptation without degrading reasoning, we propose Mix-Policy DPO, a new preference-based continual learning approach that combines on-policy stability with off-policy novelty to mitigate catastrophic forgetting during model upgrades. Deployed at production scale on Facebook and Meta AI products, RTTP delivers +91.4% improvement in tail-trend detection precision@500 and +19% query generation accuracy over industry baselines, while sustaining stable performance after multi-week online training. This work demonstrates that LLM-generated synthetic search signals, when aligned and continually updated, unlock timely trend understanding in low-traffic search environments.

cs.IR↗

arXiv2026

Scalable Machine Learning Force Fields for Macromolecular Systems Through Long-Range Aware Message Passing

Machine learning force fields (MLFFs) have revolutionized molecular simulations by providing quantum mechanical accuracy at the speed of molecular mechanical computations. However, a fundamental reliance of these models on fixed-cutoff architectures limits their applicability to macromolecular systems where long-range interactions dominate. We demonstrate that this locality constraint causes force prediction errors to scale monotonically with system size, revealing a critical architectural bottleneck. To overcome this, we establish the systematically designed MolLR25 ({Mol}ecules with {L}ong-{R}ange effect) benchmark up to 1200 atoms, generated using high-fidelity DFT, and introduce E2Former-LSR, an equivariant transformer that explicitly integrates long-range attention blocks. E2Former-LSR exhibits stable error scaling, achieves superior fidelity in capturing non-covalent decay, and maintains precision on complex protein conformations. Crucially, its efficient design provides up to 30% speedup compared to purely local models. This work validates the necessity of non-local architectures for generalizable MLFFs, enabling high-fidelity molecular dynamics for large-scale chemical and biological systems.

physics.chem-ph↗

arXiv2025

Exploring MLLM-Diffusion Information Transfer with MetaCanvas

Multimodal learning has rapidly advanced visual understanding, largely via multimodal large language models (MLLMs) that use powerful LLMs as cognitive cores. In visual generation, however, these powerful core models are typically reduced to global text encoders for diffusion models, leaving most of their reasoning and planning ability unused. This creates a gap: current multimodal LLMs can parse complex layouts, attributes, and knowledge-intensive scenes, yet struggle to generate images or videos with equally precise and structured control. We propose MetaCanvas, a lightweight framework that lets MLLMs reason and plan directly in spatial and spatiotemporal latent spaces and interface tightly with diffusion generators. We empirically implement MetaCanvas on three different diffusion backbones and evaluate it across six tasks, including text-to-image generation, text/image-to-video generation, image/video editing, and in-context video generation, each requiring precise layouts, robust attribute binding, and reasoning-intensive control. MetaCanvas consistently outperforms global-conditioning baselines, suggesting that treating MLLMs as latent-space planners is a promising direction for narrowing the gap between multimodal understanding and generation.

cs.CV↗

arXiv2025

StyleBreak: Revealing Alignment Vulnerabilities in Large Audio-Language Models via Style-Aware Audio Jailbreak

Large Audio-language Models (LAMs) have recently enabled powerful speech-based interactions by coupling audio encoders with Large Language Models (LLMs). However, the security of LAMs under adversarial attacks remains underexplored, especially through audio jailbreaks that craft malicious audio prompts to bypass alignment. Existing efforts primarily rely on converting text-based attacks into speech or applying shallow signal-level perturbations, overlooking the impact of human speech's expressive variations on LAM alignment robustness. To address this gap, we propose StyleBreak, a novel style-aware audio jailbreak framework that systematically investigates how diverse human speech attributes affect LAM alignment robustness. Specifically, StyleBreak employs a two-stage style-aware transformation pipeline that perturbs both textual content and audio to control linguistic, paralinguistic, and extralinguistic attributes. Furthermore, we develop a query-adaptive policy network that automatically searches for adversarial styles to enhance the efficiency of LAM jailbreak exploration. Extensive evaluations demonstrate that LAMs exhibit critical vulnerabilities when exposed to diverse human speech attributes. Moreover, StyleBreak achieves substantial improvements in attack effectiveness and efficiency across multiple attack paradigms, highlighting the urgent need for more robust alignment in LAMs.

cs.SD↗

arXiv2025

E2Former: An Efficient and Equivariant Transformer with Linear-Scaling Tensor Products

Equivariant Graph Neural Networks (EGNNs) have demonstrated significant success in modeling microscale systems, including those in chemistry, biology and materials science. However, EGNNs face substantial computational challenges due to the high cost of constructing edge features via spherical tensor products, making them impractical for large-scale systems. To address this limitation, we introduce E2Former, an equivariant and efficient transformer architecture that incorporates the Wigner $6j$ convolution (Wigner $6j$ Conv). By shifting the computational burden from edges to nodes, the Wigner $6j$ Conv reduces the complexity from $O(|\mathcal{E}|)$ to $ O(| \mathcal{V}|)$ while preserving both the model's expressive power and rotational equivariance. We show that this approach achieves a 7x-30x speedup compared to conventional $\mathrm{SO}(3)$ convolutions. Furthermore, our empirical results demonstrate that the derived E2Former mitigates the computational challenges of existing approaches without compromising the ability to capture detailed geometric information. This development could suggest a promising direction for scalable and efficient molecular modeling.

cs.LG↗

arXiv2025

Domain Adaptation-based Edge Computing for Cross-Conditions Fault Diagnosis

Fault diagnosis of mechanical equipment provides robust support for industrial production. It is worth noting that, the operation of mechanical equipment is accompanied by changes in factors such as speed and load, leading to significant differences in data distribution, which pose challenges for fault diagnosis. Additionally, in terms of application deployment, commonly used cloud-based fault diagnosis methods often encounter issues such as time delays and data security concerns, while common fault diagnosis methods cannot be directly applied to edge computing devices. Therefore, conducting fault diagnosis under cross-operating conditions based on edge computing holds significant research value. This paper proposes a domain-adaptation-based lightweight fault diagnosis framework tailored for edge computing scenarios. Incorporating the local maximum mean discrepancy into knowledge transfer aligns the feature distributions of different domains in a high-dimensional feature space, to discover a common feature space across domains. The acquired fault diagnosis expertise from the cloud-based deep neural network model is transferred to the lightweight edge-based model (edge model) using adaptation knowledge transfer methods. It aims to achieve accurate fault diagnosis under cross-working conditions while ensuring real-time diagnosis capabilities. We utilized the NVIDIA Jetson Xavier NX kit as the edge computing platform and conducted validation experiments on two devices. In terms of diagnostic performance, the proposed method significantly improved diagnostic accuracy, with average increases of 34.44% and 17.33% compared to existing methods, respectively.

cs.DC↗

arXiv2025

Improving Chain-of-Thought Efficiency for Autoregressive Image Generation

Autoregressive multimodal large language models have recently gained popularity for image generation, driven by advances in foundation models. To enhance alignment and detail, newer approaches employ chain-of-thought (CoT) reasoning, expanding user inputs into elaborated prompts prior to image synthesis. However, this strategy can introduce unnecessary redundancy -- a phenomenon we call visual overthinking -- which increases computational costs and can introduce details that contradict the original prompt. In this work, we explore how to generate more concise CoT sequences for more efficient image generation. We introduce ShortCoTI, a lightweight optimization framework that encourages more concise CoT while preserving output image quality. ShortCoTI rewards more concise prompts with an adaptive function that scales according to an estimated difficulty for each task. Incorporating this reward into a reinforcement learning paradigm reduces prompt reasoning length by 54% while maintaining or slightly improving quality metrics across multiple benchmarks (T2I-CompBench, GenEval). Qualitative analysis shows that our method eliminates verbose explanations and repetitive refinements, producing reasoning prompts that are both concise and semantically rich. As a result, ShortCoTI improves computational efficiency without compromising the fidelity or visual appeal of generated images.

cs.CV↗

arXiv2025

Joint Optimization and Statistical Inference for Zero-th Order Simulation Optimization

We consider stochastic optimization problems with the dual tasks of (i) effectively finding the optimizer and (ii) reliably conducting statistical inference for the optimal objective function value. We find that classical simulation optimization and stochastic optimization algorithms, despite of their fast convergence rates to the optimizer under strong convexity assumptions, may not come with a valid central limit theorem (CLT) with a vanishing bias. This non-vanishing bias can harm statistical inference and the construction of asymptotically valid confidence intervals. We fix this issue by providing a new stochastic optimization algorithm that on one hand maintains the same fast convergence rate and on the other hand permits the establishment of a valid CLT with vanishing bias. We discuss practical implementations of the proposed algorithm and conduct numerical experiments to illustrate the theoretical findings.

stat.ME↗

arXiv2025

Classifier Language Models: Unifying Sparse Finetuning and Adaptive Tokenization for Specialized Classification Tasks

Semantic text classification requires the understanding of the contextual significance of specific tokens rather than surface-level patterns or keywords (as in rule-based or statistical text classification), making large language models (LLMs) well-suited for this task. However, semantic classification applications in industry, like customer intent detection or semantic role labeling, tend to be highly specialized. They require annotation by domain experts in contrast to general-purpose corpora for pretraining. Further, they typically require high inference throughputs which limits the model size from latency and cost perspectives. Thus, for a range of specialized classification tasks, the preferred solution is to develop customized classifiers by finetuning smaller language models (e.g., mini-encoders, small language models). In this work, we develop a token-driven sparse finetuning strategy to adapt small language models to specialized classification tasks. We identify and finetune a small sensitive subset of model parameters by leveraging task-specific token constructs in the finetuning dataset, while leaving most of the pretrained weights unchanged. Unlike adapter approaches such as low rank adaptation (LoRA), we do not introduce additional parameters to the model. Our approach identifies highly relevant semantic tokens (case study in the Appendix) and outperforms end-to-end finetuning, LoRA, layer selection, and prefix tuning on five diverse semantic classification tasks. We achieve greater stability and half the training costs vs. end-to-end finetuning.

cs.LG↗

arXiv2024

JailPO: A Novel Black-box Jailbreak Framework via Preference Optimization against Aligned LLMs

Large Language Models (LLMs) aligned with human feedback have recently garnered significant attention. However, it remains vulnerable to jailbreak attacks, where adversaries manipulate prompts to induce harmful outputs. Exploring jailbreak attacks enables us to investigate the vulnerabilities of LLMs and further guides us in enhancing their security. Unfortunately, existing techniques mainly rely on handcrafted templates or generated-based optimization, posing challenges in scalability, efficiency and universality. To address these issues, we present JailPO, a novel black-box jailbreak framework to examine LLM alignment. For scalability and universality, JailPO meticulously trains attack models to automatically generate covert jailbreak prompts. Furthermore, we introduce a preference optimization-based attack method to enhance the jailbreak effectiveness, thereby improving efficiency. To analyze model vulnerabilities, we provide three flexible jailbreak patterns. Extensive experiments demonstrate that JailPO not only automates the attack process while maintaining effectiveness but also exhibits superior performance in efficiency, universality, and robustness against defenses compared to baselines. Additionally, our analysis of the three JailPO patterns reveals that attacks based on complex templates exhibit higher attack strength, whereas covert question transformations elicit riskier responses and are more likely to bypass defense mechanisms.

cs.CR↗

arXiv2024

Establishing a New Benchmark in Quantum Computational Advantage with 105-qubit Zuchongzhi 3.0 Processor

In the relentless pursuit of quantum computational advantage, we present a significant advancement with the development of Zuchongzhi 3.0. This superconducting quantum computer prototype, comprising 105 qubits, achieves high operational fidelities, with single-qubit gates, two-qubit gates, and readout fidelity at 99.90%, 99.62% and 99.18%, respectively. Our experiments with an 83-qubit, 32-cycle random circuit sampling on Zuchongzhi 3.0 highlight its superior performance, achieving one million samples in just a few hundred seconds. This task is estimated to be infeasible on the most powerful classical supercomputers, Frontier, which would require approximately $6.4\times 10^9$ years to replicate the task. This leap in processing power places the classical simulation cost six orders of magnitude beyond Google's SYC-67 and SYC-70 experiments [Nature 634, 328(2024)], firmly establishing a new benchmark in quantum computational advantage. Our work not only advances the frontiers of quantum computing but also lays the groundwork for a new era where quantum processors play an essential role in tackling sophisticated real-world challenges.

quant-ph↗

arXiv2024

Large Language Model Enhanced Machine Learning Estimators for Classification

Pre-trained large language models (LLM) have emerged as a powerful tool for simulating various scenarios and generating output given specific instructions and multimodal input. In this work, we analyze the specific use of LLM to enhance a classical supervised machine learning method for classification problems. We propose a few approaches to integrate LLM into a classical machine learning estimator to further enhance the prediction performance. We examine the performance of the proposed approaches through both standard supervised learning binary classification tasks, and a transfer learning task where the test data observe distribution changes compared to the training data. Numerical experiments using four publicly available datasets are conducted and suggest that using LLM to enhance classical machine learning estimators can provide significant improvement on prediction performance.

cs.LG↗

arXiv2024

A diverse set of two-qubit gates for spin qubits in semiconductor quantum dots

To realize large-scale quantum information processes, an ideal scheme for two-qubit operations should enable diverse operations with given hardware and physical interaction. However, for spin qubits in semiconductor quantum dots, the common two-qubit operations, including CPhase gates, SWAP gates, and CROT gates, are realized with distinct parameter regions and control waveforms, posing challenges for their simultaneous implementation. Here, taking advantage of the inherent Heisenberg interaction between spin qubits, we propose and verify a fast composite two-qubit gate scheme to extend the available two-qubit gate types as well as reduce the requirements for device properties. Apart from the formerly proposed CPhase (controlled-phase) gates and SWAP gates, theoretical results indicate that the iSWAP-family gate and Fermionic simulation (fSim) gate set are additionally available for spin qubits. Meanwhile, our gate scheme limits the parameter requirements of all essential two-qubit gates to a common J~ΔE_Z region, facilitate the simultaneous realization of them. Furthermore, we present the preliminary experimental demonstration of the composite gate scheme, observing excellent match between the measured and simulated results. With this versatile composite gate scheme, broad-spectrum two-qubit operations allow us to efficiently utilize the hardware and the underlying physics resources, helping accelerate and broaden the scope of the upcoming noise intermediate-scale quantum (NISQ) computing.

cond-mat.mes-hall↗

arXiv2024

Dynamic Gaussian Graph Operator: Learning parametric partial differential equations in arbitrary discrete mechanics problems

Deep learning methods have access to be employed for solving physical systems governed by parametric partial differential equations (PDEs) due to massive scientific data. It has been refined to operator learning that focuses on learning non-linear mapping between infinite-dimensional function spaces, offering interface from observations to solutions. However, state-of-the-art neural operators are limited to constant and uniform discretization, thereby leading to deficiency in generalization on arbitrary discretization schemes for computational domain. In this work, we propose a novel operator learning algorithm, referred to as Dynamic Gaussian Graph Operator (DGGO) that expands neural operators to learning parametric PDEs in arbitrary discrete mechanics problems. The Dynamic Gaussian Graph (DGG) kernel learns to map the observation vectors defined in general Euclidean space to metric vectors defined in high-dimensional uniform metric space. The DGG integral kernel is parameterized by Gaussian kernel weighted Riemann sum approximating and using dynamic message passing graph to depict the interrelation within the integral term. Fourier Neural Operator is selected to localize the metric vectors on spatial and frequency domains. Metric vectors are regarded as located on latent uniform domain, wherein spatial and spectral transformation offer highly regular constraints on solution space. The efficiency and robustness of DGGO are validated by applying it to solve numerical arbitrary discrete mechanics problems in comparison with mainstream neural operators. Ablation experiments are implemented to demonstrate the effectiveness of spatial transformation in the DGG kernel. The proposed method is utilized to forecast stress field of hyper-elastic material with geometrically variable void as engineering application.

cs.LG↗

arXiv2024

MPIPN: A Multi Physics-Informed PointNet for solving parametric acoustic-structure systems

Machine learning is employed for solving physical systems governed by general nonlinear partial differential equations (PDEs). However, complex multi-physics systems such as acoustic-structure coupling are often described by a series of PDEs that incorporate variable physical quantities, which are referred to as parametric systems. There are lack of strategies for solving parametric systems governed by PDEs that involve explicit and implicit quantities. In this paper, a deep learning-based Multi Physics-Informed PointNet (MPIPN) is proposed for solving parametric acoustic-structure systems. First, the MPIPN induces an enhanced point-cloud architecture that encompasses explicit physical quantities and geometric features of computational domains. Then, the MPIPN extracts local and global features of the reconstructed point-cloud as parts of solving criteria of parametric systems, respectively. Besides, implicit physical quantities are embedded by encoding techniques as another part of solving criteria. Finally, all solving criteria that characterize parametric systems are amalgamated to form distinctive sequences as the input of the MPIPN, whose outputs are solutions of systems. The proposed framework is trained by adaptive physics-informed loss functions for corresponding computational domains. The framework is generalized to deal with new parametric conditions of systems. The effectiveness of the MPIPN is validated by applying it to solve steady parametric acoustic-structure coupling systems governed by the Helmholtz equations. An ablation experiment has been implemented to demonstrate the efficacy of physics-informed impact with a minority of supervised data. The proposed method yields reasonable precision across all computational domains under constant parametric conditions and changeable combinations of parametric conditions for acoustic-structure systems.

cs.LG↗

arXiv2023

A SWAP Gate for Spin Qubits in Silicon

With one- and two-qubit gate fidelities approaching the fault-tolerance threshold for spin qubits in silicon, how to scale up the architecture and make large arrays of spin qubits become the more pressing challenges. In a scaled-up structure, qubit-to-qubit connectivity has crucial impact on gate counts of quantum error correction and general quantum algorithms. In our toolbox of quantum gates for spin qubits, SWAP gate is quite versatile: it can help solve the connectivity problem by realizing both short- and long-range spin state transfer, and act as a basic two-qubit gate, which can reduce quantum circuit depth when combined with other two-qubit gates. However, for spin qubits in silicon quantum dots, high fidelity SWAP gates have not been demonstrated due to the requirements of large circuit bandwidth and a highly adjustable ratio between the strength of the exchange coupling J and the Zeeman energy difference Delta E_z. Here we demonstrate a fast SWAP gate with a duration of ~25 ns based on quantum dots in isotopically enriched silicon, with a highly adjustable ratio between J and Delta E_z, for over two orders of magnitude in our device. We are also able to calibrate the single-qubit local phases during the SWAP gate by incorporating single-qubit gates in our circuit. By independently reading out the qubits, we probe the anti-correlations between the two spins, estimate the operation fidelity and analyze the dominant error sources for our SWAP gate. These results pave the way for high fidelity SWAP gates, and processes based on them, such as quantum communication on chip and quantum simulation by engineering the Heisenberg Hamiltonian in silicon.

cond-mat.mes-hall↗

arXiv2023

Single spin qubit geometric gate in a silicon quantum dot

Preserving qubit coherence and maintaining high-fidelity qubit control under complex noise environment is an enduring challenge for scalable quantum computing. Here we demonstrate an addressable fault-tolerant single spin qubit with an average control fidelity of 99.12% via randomized benchmarking on a silicon quantum dot device with an integrated micromagnet. Its dephasing time T2* is 1.025 us and can be enlarged to 264 us by using the Hahn echo technique, reflecting strong low-frequency noise in our system. To break through the noise limitation, we introduce geometric quantum computing to obtain high control fidelity by exploiting its noise-resilient feature. However, the control fidelities of the geometric quantum gates are lower than 99%. According to our simulation, the noise-resilient feature of geometric quantum gates is masked by the heating effect. With further optimization to alleviate the heating effect, geometric quantum computing can be a potential approach to reproducibly achieving high-fidelity qubit control in a complex noise environment.

cond-mat.mes-hall↗

arXiv2023

Correcting on-chip distortion of control pulses with silicon spin qubits

Pulse distortion, as one of the coherent error sources, hinders the characterization and control of qubits. In the semiconductor quantum dot system, the distortions on measurement pulses and control pulses disturb the experimental results, while no effective calibration procedure has yet been reported. Here, we demonstrate two different calibration methods to calibrate and correct the distortion using the two-qubit system as a detector. The two calibration methods have different correction accuracy and complexity. One is the coarse predistortion (CPD) method, with which the distortion is partly relieved. The other method is the all predistortion (APD) method, with which we measure the transfer function and significantly improve the exchange oscillation homogeneity. The two methods use the exchange oscillation homogeneity as the metric and are appropriate for any qubit that oscillates with a diabatic pulse. With the APD procedure, an arbitrary control waveform can be accurately delivered to the device, which is essential for characterizing qubits and improving gate fidelity.

cond-mat.mes-hall↗