arXiv subjects

Yufeng Zhang

Publications and source records attributed to Yufeng Zhang.

At least 37 records · Page 2Linked to original sources

arXiv2026

On-chip Multimode Opto-electronic Neural Network

Opto-electronic computing combines the complementary strengths of photonics and electronics to deliver ultrahigh computational throughput with high energy efficiency. However, its practical deployment for real-world applications has been limited by architectures that rely on delicate wavelength management or phase-sensitive coherent detection. Here, we demonstrate the first multimode opto-electronic neural network (MOENN) on a silicon-on-insulator platform. By utilizing orthogonal waveguide eigenmodes as independent information carriers, our architecture achieves robust single-wavelength computation that is inherently immune to spectral crosstalk and phase noise. The fabricated MOENN chip monolithically integrates all functional components, including input encoders, programmable mode-division fan-in/-out units, and most importantly, the nonlinear multimode activation functions. We report the system's versatility through in-situ training via a genetic algorithm, successfully resolving the nonlinear decision boundaries of a two-class dataset and achieving 92.1% accuracy on the Iris classification benchmark. Furthermore, we reconfigure the MOENN into a one-dimensional convolutional neural network, attaining an accuracy of 90.7% on the electrocardiogram-based emotion recognition task. This work establishes a new opto-electronic computing paradigm of simple control and excellent robustness, providing a compelling path toward scalable, deployable photonic intelligence.

physics.optics↗

arXiv2026

S$^2$F: Principled Hybrid Testing With Fuzzing, Symbolic Execution, and Sampling

Hybrid testing that integrates fuzzing, symbolic execution, and sampling has demonstrated superior testing efficiency compared to individual techniques. However, the state-of-the-art (SOTA) hybrid testing tools do not fully exploit the capabilities of symbolic execution and sampling in two key aspects. First, the SOTA hybrid testing tools employ tailored symbolic execution engines that tend to over-prune branches, leading to considerable time wasted waiting for seeds from the fuzzer and missing opportunities to discover crashes. Second, existing methods do not apply sampling to the appropriate branches and therefore cannot utilize the full capability of sampling. To address these two limitations, we propose a novel hybrid testing architecture that combines the precision of conventional symbolic execution with the scalability of tailored symbolic execution engines. Based on this architecture, we propose several principles for combining fuzzing, symbolic execution, and sampling. We implement our method in a hybrid testing tool S$^2$F. To evaluate its effectiveness, we conduct extensive experiments on 15 real-world programs. Experimental results demonstrate that S$^2$F outperforms the SOTA tool, achieving an average improvement of 6.14% in edge coverage and 32.6% in discovered crashes. Notably, our tool uncovers three previously unknown crashes in real-world programs.

cs.SE↗

arXiv2025

MCOP: Multi-UAV Collaborative Occupancy Prediction

Unmanned Aerial Vehicle (UAV) swarm systems necessitate efficient collaborative perception mechanisms for diverse operational scenarios. Current Bird's Eye View (BEV)-based approaches exhibit two main limitations: bounding-box representations fail to capture complete semantic and geometric information of the scene, and their performance significantly degrades when encountering undefined or occluded objects. To address these limitations, we propose a novel multi-UAV collaborative occupancy prediction framework. Our framework effectively preserves 3D spatial structures and semantics through integrating a Spatial-Aware Feature Encoder and Cross-Agent Feature Integration. To enhance efficiency, we further introduce Altitude-Aware Feature Reduction to compactly represent scene information, along with a Dual-Mask Perceptual Guidance mechanism to adaptively select features and reduce communication overhead. Due to the absence of suitable benchmark datasets, we extend three datasets for evaluation: two virtual datasets (Air-to-Pred-Occ and UAV3D-Occ) and one real-world dataset (GauUScene-Occ). Experiments results demonstrate that our method achieves state-of-the-art accuracy, significantly outperforming existing collaborative methods while reducing communication overhead to only a fraction of previous approaches.

cs.CV↗

arXiv2025

Protein as a Second Language for LLMs

Deciphering the function of unseen protein sequences is a fundamental challenge with broad scientific impact, yet most existing methods depend on task-specific adapters or large-scale supervised fine-tuning. We introduce the "Protein-as-Second-Language" framework, which reformulates amino-acid sequences as sentences in a novel symbolic language that large language models can interpret through contextual exemplars. Our approach adaptively constructs sequence-question-answer triples that reveal functional cues in a zero-shot setting, without any further training. To support this process, we curate a bilingual corpus of 79,926 protein-QA instances spanning attribute prediction, descriptive understanding, and extended reasoning. Empirically, our method delivers consistent gains across diverse open-source LLMs and GPT-4, achieving up to 17.2% ROUGE-L improvement (average +7%) and even surpassing fine-tuned protein-specific language models. These results highlight that generic LLMs, when guided with protein-as-language cues, can outperform domain-specialized models, offering a scalable pathway for protein understanding in foundation models.

cs.LG↗

arXiv2025

CPT: Competence-progressive Training Strategy for Few-shot Node Classification

Graph Neural Networks (GNNs) have made significant advancements in node classification, but their success relies on sufficient labeled nodes per class in the training data. Real-world graph data often exhibits a long-tail distribution with sparse labels, emphasizing the importance of GNNs' ability in few-shot node classification, which entails categorizing nodes with limited data. Traditional episodic meta-learning approaches have shown promise in this domain, but they face an inherent limitation: it might lead the model to converge to suboptimal solutions because of random and uniform task assignment, ignoring task difficulty levels. This could lead the meta-learner to face complex tasks too soon, hindering proper learning. Ideally, the meta-learner should start with simple concepts and advance to more complex ones, like human learning. So, we introduce CPT, a novel two-stage curriculum learning method that aligns task difficulty with the meta-learner's progressive competence, enhancing overall performance. Specifically, in CPT's initial stage, the focus is on simpler tasks, fostering foundational skills for engaging with complex tasks later. Importantly, the second stage dynamically adjusts task difficulty based on the meta-learner's growing competence, aiming for optimal knowledge acquisition. Extensive experiments on popular node classification datasets demonstrate significant improvements of our strategy over existing methods.

cs.LG↗

arXiv2025

ABC: Adaptive BayesNet Structure Learning for Computational Scalable Multi-task Image Compression

Neural Image Compression (NIC) has revolutionized image compression with its superior rate-distortion performance and multi-task capabilities, supporting both human visual perception and machine vision tasks. However, its widespread adoption is hindered by substantial computational demands. While existing approaches attempt to address this challenge through module-specific optimizations or pre-defined complexity levels, they lack comprehensive control over computational complexity. We present ABC (Adaptive BayesNet structure learning for computational scalable multi-task image Compression), a novel, comprehensive framework that achieves computational scalability across all NIC components through Bayesian network (BayesNet) structure learning. ABC introduces three key innovations: (i) a heterogeneous bipartite BayesNet (inter-node structure) for managing neural backbone computations; (ii) a homogeneous multipartite BayesNet (intra-node structure) for optimizing autoregressive unit processing; and (iii) an adaptive control module that dynamically adjusts the BayesNet structure based on device capabilities, input data complexity, and downstream task requirements. Experiments demonstrate that ABC enables full computational scalability with better complexity adaptivity and broader complexity control span, while maintaining competitive compression performance. Furthermore, the framework's versatility allows integration with various NIC architectures that employ BayesNet representations, making it a robust solution for ensuring computational scalability in NIC applications. Code is available in https://github.com/worldlife123/cbench_BaSIC.

eess.IV↗

arXiv2025

Direct tensor processing with coherent light

Tensor processing is the cornerstone of modern technological advancements, powering critical applications in data analytics and artificial intelligence. While optical computing offers exceptional advantages in bandwidth, parallelism, and energy efficiency, existing methods optimized for scalar operations struggle to efficiently handle tensor-based tasks, limiting their applicability in complex applications, such as neural networks. Here, we report Parallel Optical Matrix Matrix Multiplication (POMMM), a novel paradigm that enables fully parallel tensor processing through a single coherent light propagation. This approach addresses key limitations of current optical methods, scaling the performance with data dimension, while improving theoretical computational power and efficiency. We demonstrate its high consistency with GPU based matrix matrix multiplication across both real-valued and complex valued domains. Moreover, we showcase its adaptability, scalability, and versatility in tensor processing applications such as convolutional and vision transformer neural networks. Furthermore, we analyse the theoretical compatibility and efficiency of POMMM in relation to existing optical computing paradigms, highlighting its potential to outperform current state-of-the-art methods. By enabling a variety of computational tasks and supporting multi2 wavelength and large-scale expansion, POMMM provides a scalable, high-efficient foundation for advancing next-generation optical computing.

physics.optics↗

arXiv2025

FullStack Bench: Evaluating LLMs as Full Stack Coders

As the capabilities of code large language models (LLMs) continue to expand, their applications across diverse code intelligence domains are rapidly increasing. However, most existing datasets only evaluate limited application domains. To address this gap, we have developed a comprehensive code evaluation dataset FullStack Bench focusing on full-stack programming, which encompasses a wide range of application domains (e.g., basic programming, data analysis, software engineering, mathematics, and machine learning). Besides, to assess multilingual programming capabilities, in FullStack Bench, we design real-world instructions and corresponding unit test cases from 16 widely-used programming languages to reflect real-world usage scenarios rather than simple translations. Moreover, we also release an effective code sandbox execution tool (i.e., SandboxFusion) supporting various programming languages and packages to evaluate the performance of our FullStack Bench efficiently. Comprehensive experimental results on our FullStack Bench demonstrate the necessity and effectiveness of our FullStack Bench and SandboxFusion.

cs.AI↗

arXiv2025

Reward-Augmented Data Enhances Direct Preference Alignment of LLMs

Preference alignment in Large Language Models (LLMs) has significantly improved their ability to adhere to human instructions and intentions. However, existing direct alignment algorithms primarily focus on relative preferences and often overlook the qualitative aspects of responses, despite having access to preference data that includes reward scores from judge models during AI feedback. Striving to maximize the implicit reward gap between the chosen and the slightly inferior rejected responses can cause overfitting and unnecessary unlearning of the high-quality rejected responses. The unawareness of the reward scores also drives the LLM to indiscriminately favor the low-quality chosen responses and fail to generalize to optimal responses that are sparse in data. To overcome these shortcomings, our study introduces reward-conditioned LLM policies that discern and learn from the entire spectrum of response quality within the dataset, helping extrapolate to more optimal regions. We propose an effective yet simple data relabeling method that conditions the preference pairs on quality scores to construct a reward-augmented dataset. The experiments across various benchmarks and diverse models demonstrate that our approach consistently boosts DPO by a considerable margin. Through comprehensive ablation studies, we demonstrate that our method not only maximizes the utility of preference data but also mitigates the issue of unlearning, demonstrating its broad effectiveness beyond mere data expansion. Our code is available at https://github.com/shenao-zhang/reward-augmented-preference.

cs.LG↗

arXiv2025

Strong correlation between $H$-linear magnetoresistance and strange metal in FeSe superconductor

In strange metals, a strong and anomalous scattering effect exists and increases linearly with temperature. In FeSe, we observed that the temperature dependence of resistivity exhibits non-Fermi liquid behavior in two regions below and above a critical pressure, $p_\text{c}$$\sim$2 GPa. As pressure increases, a transition from quadratic to nonsaturating magnetoresistance is observed, with a distinct crossover between these two behaviors indicated by $B^{*}$ in the derivative analysis. After subtracting the quadratic term from the magnetoresistance, the residual magnetoresistance clearly exhibits an $H$-linear behavior. Additionally, two segments of $H$-inear magnetoresistance appear with increasing pressure, each arising from distinct origins. Notably, the two $H$-linear magnetoresistances exist and develop within the strange metal states. These results suggest that $H$-linear magnetoresistance is in strong correlation with the strange metal state, which may affect superconductivity in FeSe under pressure. Our study provides valuable insights into the strange metal state and clues for underlying unconventional superconductivity.

cond-mat.supr-con↗

arXiv2025

Enhancing Chain of Thought Prompting in Large Language Models via Reasoning Patterns

Chain of Thought (CoT) prompting can encourage language models to engage in multi-step logical reasoning. The quality of the provided demonstrations significantly influences the success of downstream inference tasks. Current unsupervised CoT methods primarily select examples based on the semantics of the questions, which can introduce noise and lack interpretability. In this paper, we propose leveraging reasoning patterns to enhance CoT prompting effectiveness. Reasoning patterns represent the process by which language models arrive at their final results. By utilizing prior knowledge and prompt-based methods from large models, we first construct task-specific pattern sets. We then select diverse demonstrations based on different reasoning patterns. This approach not only mitigates the impact of noise but also provides explicit interpretability to help us understand the mechanisms of CoT. Extensive experiments demonstrate that our method is more robust and consistently leads to improvements across various reasoning tasks.

cs.CL↗

arXiv2024

Seed-CTS: Unleashing the Power of Tree Search for Superior Performance in Competitive Coding Tasks

Competition-level code generation tasks pose significant challenges for current state-of-the-art large language models (LLMs). For example, on the LiveCodeBench-Hard dataset, models such as O1-Mini and O1-Preview achieve pass@1 rates of only 0.366 and 0.143, respectively. While tree search techniques have proven effective in domains like mathematics and general coding, their potential in competition-level code generation remains under-explored. In this work, we propose a novel token-level tree search method specifically designed for code generation. Leveraging Qwen2.5-Coder-32B-Instruct, our approach achieves a pass rate of 0.305 on LiveCodeBench-Hard, surpassing the pass@100 performance of GPT4o-0513 (0.245). Furthermore, by integrating Chain-of-Thought (CoT) prompting, we improve our method's performance to 0.351, approaching O1-Mini's pass@1 rate. To ensure reproducibility, we report the average number of generations required per problem by our tree search method on the test set. Our findings underscore the potential of tree search to significantly enhance performance on competition-level code generation tasks. This opens up new possibilities for large-scale synthesis of challenging code problems supervised fine-tuning (SFT) data, advancing competition-level code generation tasks.

cs.AI↗

arXiv2024

Detecting Emotional Incongruity of Sarcasm by Commonsense Reasoning

This paper focuses on sarcasm detection, which aims to identify whether given statements convey criticism, mockery, or other negative sentiment opposite to the literal meaning. To detect sarcasm, humans often require a comprehensive understanding of the semantics in the statement and even resort to external commonsense to infer the fine-grained incongruity. However, existing methods lack commonsense inferential ability when they face complex real-world scenarios, leading to unsatisfactory performance. To address this problem, we propose a novel framework for sarcasm detection, which conducts incongruity reasoning based on commonsense augmentation, called EICR. Concretely, we first employ retrieval-augmented large language models to supplement the missing but indispensable commonsense background knowledge. To capture complex contextual associations, we construct a dependency graph and obtain the optimized topology via graph refinement. We further introduce an adaptive reasoning skeleton that integrates prior rules to extract sentiment-inconsistent subgraphs explicitly. To eliminate the possible spurious relations between words and labels, we employ adversarial contrastive learning to enhance the robustness of the detector. Experiments conducted on five datasets demonstrate the effectiveness of EICR.

cs.CL↗

arXiv2024

DavIR: Data Selection via Implicit Reward for Large Language Models

We introduce DavIR, a model-based data selection method for post-training Large Language Models. DavIR generalizes Reducible Holdout Loss to core-set selection problem of causal language modeling, and quantifies the learnability of a given datum with respect to a pre-trained LLM based on relative reduction in loss during fine-tuning, a metric we show to be closely related to the implicit reward model described in Direct Preference Optimization (DPO). We show that 6% of Alpaca dataset selected with DavIR can steer both the LLaMA and Gemma model family to produce superior performance compared to the same models trained on the full 52K dataset. We also show that Alpaca dataset compressed with DavIR can be combined with GSM8K dataset to effectively balance open-domain freeform QA and mathematical reasoning capabilities. Finally, we apply the DavIR objective to DPO and develop a normalized DavIR-DPO objective which improves alignment performance of Zephyr-7B-SFT model by 8% (relative) on AlpacaEval, compared against training on vanilla DPO objective.

cs.LG↗

arXiv2024

Long-time asymptotics for a complex cubic Camassa-Holm equation

In this paper, we investigate the Cauchy problem of the following complex cubic Camassa-Holm (ccCH) equation $$m_{t}=b u_{x}+\frac{1}{2}\left[m\left(|u|^{2}-\left|u_{x}\right|^{2}\right)\right]_{x}-\frac{1}{2} m\left(u \bar{u}_{x}-u_{x} \bar{u}\right), \quad m=u-u_{x x},$$ where $b>0$ is an arbitrary positive real constant. Long-time asymptotics of the equation is obtained through the $\bar{\partial}$-steepest descent method. Firstly, based on the spectral analysis of the Lax pair and scattering matrix, the solution of the equation is able to be constructed via solving the corresponding Riemann-Hilbert (RH) problem. Then, we present different long time asymptotic expansions of the solution $u(y,t)$ in different space-time solitonic regions of $ξ=y/t$. The half-plane ${(y,t):-\infty 0}$ is divided into four asymptotic regions: $ξ\in(-\infty,-1)$, $ξ\in (-1,0)$, $ξ\in (0,\frac{1}{8})$ and $ξ\in (\frac{1}{8},+\infty)$. When $ξ$ falls in $(-\infty,-1)\cup (\frac{1}{8},+\infty)$, no stationary phase point of the phase function $θ(z)$ exists on the jump profile in the space-time region. In this case, corresponding asymptotic approximations can be characterized with an $N(Λ)$-solitons with diverse residual error order $O(t^{-1+2\varepsilon})$. There are four stationary phase points and eight stationary phase points on the jump curve as $ξ\in (-1,0)$ and $ξ\in (0,\frac{1}{8})$, respectively. The corresponding asymptotic form is accompanied by a residual error order $O(t^{-\frac{3}{4}})$.

math.AP↗

arXiv2024

Bulk and surface Dirac states accompanied by two superconducting domes in FeSe-based superconductors

Recent investigations of FeSe-based superconductors have revealed the presence of two superconducting domes, and suggest possible distinct pairing mechanisms. Two superconducting domes are commonly found in unconventional superconductors and exhibit unique normal states and electronic structures. In this study, we conducted electromagnetic transport measurements to establish a complete phase diagram, successfully observing the two superconducting domes in FeSe$_{1-x}$S$_x$ (0 $\le x \le$ 0.25) and FeSe$_{1-x}$Te$_x$ (0 $\le x \le$ 1) superconductors. The normal state resistivity on SC1 shows the strange metal state, with a power exponent approximately equal to 1 ($ρ(T)\propto T^n$ with $n\sim 1$), whereas the exponent on SC2 is less than 1. A bulk Dirac state observed on SC1, completely synchronized with the strange metal behavior, indicating a close relationship between them. While a topological surface Dirac state is witnessed on SC2, and undergoes a sign change near the pure nematic quantum critical point. The evolution of the Dirac states indicates that the appearance of the two superconducting domes may originate from the Fermi surface reconstruction. Our findings highlight distinct Dirac states and normal state resistivity across the two superconducting domes, providing convincing evidence for the existence of the two different pairing mechanisms in FeSe-based superconductors.

cond-mat.supr-con↗

arXiv2024

A Mean-Field Analysis of Neural Stochastic Gradient Descent-Ascent for Functional Minimax Optimization

This paper studies minimax optimization problems defined over infinite-dimensional function classes of overparameterized two-layer neural networks. In particular, we consider the minimax optimization problem stemming from estimating linear functional equations defined by conditional expectations, where the objective functions are quadratic in the functional spaces. We address (i) the convergence of the stochastic gradient descent-ascent algorithm and (ii) the representation learning of the neural networks. We establish convergence under the mean-field regime by considering the continuous-time and infinite-width limit of the optimization dynamics. Under this regime, the stochastic gradient descent-ascent corresponds to a Wasserstein gradient flow over the space of probability measures defined over the space of neural network parameters. We prove that the Wasserstein gradient flow converges globally to a stationary point of the minimax objective at a $O(T^{-1} + α^{-1})$ sublinear rate, and additionally finds the solution to the functional equation when the regularizer of the minimax objective is strongly convex. Here $T$ denotes the time and $α$ is a scaling parameter of the neural networks. In terms of representation learning, our results show that the feature representation induced by the neural networks is allowed to deviate from the initial one by the magnitude of $O(α^{-1})$, measured in terms of the Wasserstein distance. Finally, we apply our general results to concrete examples including policy evaluation, nonparametric instrumental variable regression, asset pricing, and adversarial Riesz representer estimation.

cs.LG↗

arXiv2024

A class of extended high-dimensional nonisospectral KdV hierarchies and symmetry

We construct a new class of N-dimensional Lie algebras and apply them to integrable systems. In this paper, we obtain a nonisospectral KdV integrable hierarchy by introducing a nonisospectral spectral problem. Then, a coupled nonisospectral KdV hierarchy is deduced by means of the corresponding higher-dimensional loop algebra. It follows that the K symmetries, τ symmetries and their Lie algebra of the coupled nonisospectral KdV hierarchy are investigated. The bi-Hamiltonian structures of the both resulting hierarchies are derived by using the trace identity. Finally, we derive a multi-component nonisospectral KdV hierarchy related to the N-dimensional loop algebra, which generalizes the coupled results to an arbitrary number of components.

math-ph↗