arXiv subjects

Yiquan Wang

Publications and source records attributed to Yiquan Wang.

17 recordsLinked to original sources

arXiv2026

Spectral analysis of protein backbone geometry reveals abrupt helix--coil boundaries

The boundaries of cooperative helix--coil transitions influence protein allostery and conformational dynamics, yet the persistent one-to-two-residue ambiguity in their assignment remains poorly characterized. We apply the discrete Hasimoto map to translate three-dimensional C$_α$ backbone geometry into a one-dimensional discrete nonlinear Schrödinger effective potential and analyze its spatial-frequency structure. Helical segments appear as near-integrable, low-entropy states whose spectral power concentrates at the zero-frequency mode, whereas coil regions show broadband fluctuations. A pointwise integrability residual and a windowed spectral entropy separate the two phases with ROC AUC values of 0.783 and 0.715, and their combination reaches 0.803, while combining the residual instead with a low-frequency energy ratio reaches 0.815. Across 1\,986 proteins and 19\,148 of 21\,107 fitted helix--coil boundaries the transition is abrupt, with a median sigmoid width of 0.145 residues that measures the steepness of a single-step discrete jump rather than a literal sub-residue distance; the transition is directionally asymmetric, with helix exits sharper than entries. Across the full dataset every C$_α$ geometry-based assignment, including DSSP-calibrated P-SEA and both spectral probes, loses agreement with the DSSP hydrogen-bond reference most acutely at these boundaries, indicating that the assignment ambiguity is a general feature of C$_α$ geometry rather than any single algorithm. The windowed spectral probe is subject to a Gabor resolution limit and is therefore outperformed by the pointwise probe, which attains the lattice-limited resolution.

q-bio.BM↗

arXiv2026

Descriptive power and predictive limits of a discrete Hasimoto--DNLS model of protein backbone structure

Determining 3D protein structure from sequence remains a fundamental biophysical challenge. The C$_α$ backbone's discrete Frenet geometry maps, via a Hasimoto transform, to a complex scalar field $ψ=κ\,e^{i\sumτ}$ obeying a discrete nonlinear Schrödinger equation (DNLS), whose solitons reproduce secondary-structure motifs. Whether this compact mapping extends to a predictive folding framework remains open. We derive an exact closed-form decomposition of the DNLS effective potential $V_{\text{eff}}=V_{\text{re}}+iV_{\text{im}}$ via curvature ratios and torsion angles, validated to machine precision across 856 non-redundant proteins. Our analysis identifies three structural barriers to forward prediction: (i)~$V_{\text{im}}$ encodes chirality via the odd symmetry of $\sinτ$; its magnitude is ${\sim}31\%$ of the real part, and neglecting it causes a $2^N$ degeneracy; (ii)~$V_{\text{re}}$ is determined mostly (${\sim}95\%$) by local geometry, leaving explicit sequence dependence below ${\sim}5\%$ of variance; and (iii)~self-consistent field iterations fail to recover native structures (mean RMSD $= 13.1$\,Å) even with hydrogen-bond terms, yielding zero torsion correlations. Conversely, the DNLS dispersion relation residual serves as a geometric order parameter for $α$-helices (ROC AUC $= 0.72$), identifying where the backbone best approximates an integrable system. Thus, the Hasimoto map functions as a kinematic identity, not a dynamical governing equation. Obstacles to \textit{ab initio} prediction stem from the purely local, real-potential reduction built upon it, rather than the lossless map itself.

q-bio.BM↗

arXiv2026

StreamKL: Fast and Memory-Efficient KL Divergence for Boosting Attention Distillation

Attention distillation, which trains one attention distribution to match another by minimizing their Kullback-Leibler (KL) divergence, is widely used in knowledge distillation, model compression, continual learning, and sparse-attention LLM training. However, existing approaches materialize both attention distributions before computing the KL reduction, incurring $O(N_QN_K)$ memory and IO costs that become prohibitive at long context lengths. We present StreamKL, the first fused GPU primitive for attention KL divergence that eliminates this quadratic materialization. StreamKL derives a novel online formulation for the coupled two-distribution KL reduction, enabling a single one-pass forward kernel that streams query-key tiles through on-chip SRAM. For the backward pass, StreamKL recomputes attention probabilities tile-by-tile, avoiding storage of quadratic intermediates. We further design and implement efficient GPU kernels with dedicated optimizations. Experiments show StreamKL delivers up to $43\times$ and $14\times$ speedups over baseline methods in the forward and backward passes, respectively. Most importantly, StreamKL reduces the extra HBM footprint of attention distillation from $O(N_QN_K)$ to $O(1)$, enabling long-context distillation on a single GPU.

cs.LG↗

arXiv2026

CODO: An Automated Compiler for Comprehensive Dataflow Optimization

FPGAs are well-suited for dataflow architectures that process data in a streaming or pipelined manner, thus satisfying the high computational and communication demands of emerging applications. However, manually implementing an efficient dataflow architecture for large-scale applications is still challenging, even for specialists who use high-level synthesis (HLS) to simplify FPGA programming. To address this, we introduce CODO, an automated compiler that generates feasible and efficient dataflow accelerators on FPGAs. CODO features a systematic method for detecting and eliminating both coarse-grained and fine-grained dataflow violations. Building on this, CODO performs both on- and off-chip data movement optimizations to maximize transfer efficiency. To guarantee a higher design quality, CODO performs automatic scheduling to generate high-performance dataflow accelerators, ensuring a balanced performance-resource trade-off. Synthesis results show that CODO delivers $1.45\times$ to $4.52\times$ latency speedups on typical computation kernels and $3.7\times$ to $33.8\times$ speedups on DNN models compared to SOTA frameworks. In on-board evaluations, CODO achieves $7.3\times$ average speedup on CNN models and $2.07\times$ average speedup on the GPT-2 model over SOTA frameworks. The compiler is open-sourced at https://github.com/sjtu-zhao-lab/codo-artifact.

cs.AR↗

arXiv2026

Piecewise integrability of the discrete Hasimoto map for analytic prediction and design of helical peptides

The representation of protein backbone geometry through the discrete nonlinear Schrödinger equation provides a theoretical connection between biological structure and integrable systems. Although the global application of this framework is constrained by chiral degeneracies and non-local interactions, helical peptides can be modeled as piecewise integrable systems where the discrete Hasimoto map remains applicable within specific geometric boundaries. We delineate these boundaries through an analytic mapping $(ϕ,ψ) \rightarrow (κ,τ)$ between biochemical dihedral angles and Frenet frame parameters for 50 helical peptide chains. This transformation is globally information-preserving but ill-conditioned within the helical basin (median Jacobian condition number 31), suggesting chiral information loss arises primarily from local coordinate compression rather than topological singularities. Using a local integrability error $E[n]$ derived from the discrete dispersion relation, we show deviations from integrability are driven predominantly by torsion non-uniformity, while curvature remains rigid. This metric identifies integrable islands where the analytic dispersion relation predicts backbone coordinates with sub-angstrom accuracy (median RMSD 0.77\,Å), enabling a segmentation strategy that isolates structural defects and trims non-integrable terminal fraying. Evaluating only these integrable islands, the dispersion relation extracts high-accuracy structural cores for 88\% of the dataset. Inverse backbone design is feasible within a defined integrability zone where the design constraint reduces essentially to controlling torsion uniformity. These findings advance the Hasimoto formalism from a qualitative descriptor toward a precise quantitative framework for analyzing and designing local protein geometry within the limits of piecewise integrability.

q-bio.BM↗

arXiv2026

AI Driven Discovery of Bio Ecological Mediation in Cascading Heatwave Risks

Compound heatwaves increasingly trigger complex cascading failures that propagate through interconnected physical and human systems, yet the fragmentation of disciplinary knowledge hinders the comprehensive mapping of these systemic risk topologies. This study introduces the Heatwave Discovery Agent HeDA as an autonomous scientific synthesis framework designed to bridge cognitive gaps by constructing a high fidelity knowledge graph from 8,111 academic publications. By structuring 70,297 evidence nodes, the system exhibits enhanced inferential fidelity in capturing long tail risk mechanisms and achieves a significant accuracy margin compared to standard foundation models including GPT 5.2 and Claude Sonnet 4.5 in complex reasoning tasks. The resulting topological analysis reveals a critical bio ecological mediation effect where biological systems function as the primary non linear amplifiers of thermal stress that transform physical meteorological hazards into systemic socioeconomic losses. We further identify latent functional couplings between theoretically distinct sectors such as the heat induced synchronization of power grid failures and emergency medical capacity saturation. These findings elucidate the dynamics of compound climate risks and provide an empirical basis for shifting adaptation strategies from static sectoral defense to dynamic cross system resilience.

cs.AI↗

arXiv2026

Crypto-ncRNA: a bio-inspired post-quantum cryptographic primitive exploiting RNA folding complexity

The imminent realization of fault-tolerant quantum computing precipitates a systemic collapse of classical public-key infrastructure and necessitates an urgent transition to post-quantum cryptography. However, current standardization efforts predominantly rely on structured mathematical problems that may remain vulnerable to unforeseen algorithmic breakthroughs, highlighting a critical need for fundamentally orthogonal security paradigms. Here, we introduce \emph{Crypto-ncRNA} as a biophysically inspired cryptographic primitive that exploits the thermodynamic complexity of non-coding RNA folding as a computational work-factor amplifier. By leveraging the rugged energy landscape inherent to RNA secondary structure prediction, a problem intractable to rapid inversion, we establish a security foundation independent of conventional number-theoretic assumptions. We validate this approach by mapping the folding problem to a Quadratic Unconstrained Binary Optimization model and demonstrate theoretical resilience against quantum optimization attacks including the Quantum Approximate Optimization Algorithm. Functioning as a symmetric key encapsulation and derivation primitive dependent on pre-shared seeds, Crypto-ncRNA achieves throughputs competitive with software-based Advanced Encryption Standard implementations. By utilizing the generated high-entropy keys within a standard stream cipher framework, it exhibits ciphertext entropy that satisfies rigorous NIST SP 800-22 statistical standards. These findings not only articulate a novel bio-computational pathway for cryptographic defense but also provide a rigorous algorithmic blueprint for future physical realization, demonstrating that the thermodynamic complexity of biological systems offers a robust and physically grounded frontier for securing digital infrastructure in the post-quantum era.

cs.CR↗

arXiv2026

Construction and educational application of a linguistically grounded dependency treebank for Uyghur

Developing effective educational technologies for low-resource agglutinative languages like Uyghur is often hindered by the mismatch between existing annotation frameworks and specific grammatical structures. To address this challenge, this study introduces the Modern Uyghur Dependency Treebank (MUDT), a linguistically grounded annotation framework specifically designed to capture the agglutinative complexity of Uyghur, including zero copula constructions and fine-grained case marking. Utilizing a hybrid pipeline that combines Large Language Model pre-annotation with rigorous human correction, a high-quality treebank consisting of 3,456 sentences was constructed. Intrinsic structural evaluation reveals that MUDT significantly improves dependency projectivity by reducing the crossing-arc rate from 7.35\% in the Universal Dependencies standard to 0.06\%. Extrinsic parsing experiments using UDPipe and Stanza further demonstrate that models trained on MUDT achieve superior in-domain accuracy and cross-domain generalization compared to UD-based baselines. To validate the practical utility of this computational resource, an AI-assisted grammar tutoring system was developed to translate MUDT-based syntactic analyses into interpretable pedagogical feedback. A controlled experiment involving 35 second-language learners indicated that students receiving syntax-aware feedback achieved significantly higher learning gains compared to those in a control group. These findings establish MUDT as a robust foundation for syntactic analysis and underscore the critical role of linguistically informed natural language processing resources in bridging the gap between computational models and the cognitive needs of second-language learners.

cs.CL↗

arXiv2025

Diffusion Models at the Drug Discovery Frontier: A Review on Generating Small Molecules versus Therapeutic Peptides

Diffusion models have emerged as a leading framework in generative modeling, poised to transform the traditionally slow and costly process of drug discovery. This review provides a systematic comparison of their application in designing two principal therapeutic modalities: small molecules and therapeutic peptides. We dissect how the unified framework of iterative denoising is adapted to the distinct molecular representations, chemical spaces, and design objectives of each modality. For small molecules, these models excel at structure-based design, generating novel, pocket-fitting ligands with desired physicochemical properties, yet face the critical hurdle of ensuring chemical synthesizability. Conversely, for therapeutic peptides, the focus shifts to generating functional sequences and designing de novo structures, where the primary challenges are achieving biological stability against proteolysis, ensuring proper folding, and minimizing immunogenicity. Despite these distinct challenges, both domains face shared hurdles: the scarcity of high-quality experimental data, the reliance on inaccurate scoring functions for validation, and the crucial need for experimental validation. We conclude that the full potential of diffusion models will be unlocked by bridging these modality-specific gaps and integrating them into automated, closed-loop Design-Build-Test-Learn (DBTL) platforms, thereby shifting the paradigm from mere chemical exploration to the on-demand engineering of novel~therapeutics.

cs.LG↗

arXiv2025

Octopus Inspired Optimization (OIO): A Hierarchical Framework for Navigating Protein Fitness Landscapes

Navigating vast, rugged biological fitness landscapes to discover high-value functional patterns-such as optimal protein sequences-is a central challenge in health informatics. However, conventional algorithms often struggle with the exploration-exploitation dilemma, failing to synergize global search with deep local refinement, which leads to entrapment in suboptimal solutions. To overcome this barrier, we introduce Octopus Inspired Optimization (OIO), a novel hierarchical metaheuristic that mimics the octopus's unique neural architecture to intrinsically unify centralized global exploration and parallelized local exploitation. We validated OIO on a real-world protein engineering benchmark, where it surpassed 15 competing metaheuristics. This success is underpinned by OIO's architectural suitability for protein-like landscapes, confirmed by its top ranking on the NK-Landscape benchmark, and its powerful optimization engine, demonstrated by its first-place performance on the gold-standard CEC2022 benchmark. OIO thus provides a robust, nature-inspired computational tool for complex optimization problems in drug discovery and personalized medicine.

cs.NE↗

arXiv2025

Fractal Signatures: Securing AI-Generated Pollock-Style Art via Intrinsic Watermarking and Blockchain

The digital art market faces unprecedented challenges in authenticity verification and copyright protection. This study introduces an integrated framework to address these issues by combining neural style transfer, fractal analysis, and blockchain technology. We generate abstract artworks inspired by Jackson Pollock, using their inherent mathematical complexity to create robust, imperceptible watermarks. Our method embeds these watermarks, derived from fractal and turbulence features, directly into the artwork's structure. This approach is then secured by linking the watermark to NFT metadata, ensuring immutable proof of ownership. Rigorous testing shows our feature-based watermarking achieves a 76.2% average detection rate against common attacks, significantly outperforming traditional methods (27.8-44.0%). This work offers a practical solution for digital artists and collectors, enhancing security and trust in the digital art ecosystem.

cs.CV↗

arXiv2025

On Reducing the Execution Latency of Superconducting Quantum Processors via Quantum Job Scheduling

Quantum computing has gained considerable attention, especially after the arrival of the Noisy Intermediate-Scale Quantum (NISQ) era. Quantum processors and cloud services have been made world-wide increasingly available. Unfortunately, jobs on existing quantum processors are often executed in series, and the workload could be heavy to the processor. Typically, one has to wait for hours or even longer to obtain the result of a single quantum job on public quantum cloud due to long queue time. In fact, as the scale grows, the qubit utilization rate of the serial execution mode will further diminish, causing the waste of quantum resources. In this paper, to our best knowledge for the first time, the Quantum Job Scheduling Problem (QJSP) is formulated and introduced, and we accordingly aim to improve the utility efficiency of quantum resources. Specifically, a noise-aware quantum job scheduler (NAQJS) concerning the circuit width, number of measurement shots, and submission time of quantum jobs is proposed to reduce the execution latency. We conduct extensive experiments on a simulated Qiskit noise model, as well as on the Xiaohong (from QuantumCTek) superconducting quantum processor. Numerical results show the effectiveness in both the QPU time and turnaround time.

cs.AR↗

arXiv2025

STGCN-LSTM for Olympic Medal Prediction: Dynamic Power Modeling and Causal Policy Optimization

This paper proposes a novel hybrid model, STGCN-LSTM, to forecast Olympic medal distributions by integrating the spatio-temporal relationships among countries and the long-term dependencies of national performance. The Spatial-Temporal Graph Convolution Network (STGCN) captures geographic and interactive factors-such as coaching exchange and socio-economic links-while the Long Short-Term Memory (LSTM) module models historical trends in medal counts, economic data, and demographics. To address zero-inflated outputs (i.e., the disparity between countries that consistently yield wins and those never having won medals), a Zero-Inflated Compound Poisson (ZICP) framework is incorporated to separate random zeros from structural zeros, providing a clearer view of potential breakthrough performances. Validation includes historical backtracking, policy shock simulations, and causal inference checks, confirming the robustness of the proposed method. Results shed light on the influence of coaching mobility, event specialization, and strategic investment on medal forecasts, offering a data-driven foundation for optimizing sports policies and resource allocation in diverse Olympic contexts.

cs.LG↗

arXiv2024

Research on Travel Route Planing Problems Based on Greedy Algorithm

The route planning problem based on the greedy algorithm represents a method of identifying the optimal or near-optimal route between a given start point and end point. In this paper, the PCA method is employed initially to downscale the city evaluation indexes, extract the key principal components, and then downscale the data using the KMO and TOPSIS algorithms, all of which are based on the MindSpore framework. Secondly, for the dataset that does not pass the KMO test, the entropy weight method and TOPSIS method will be employed for comprehensive evaluation. Finally, a route planning algorithm is proposed and optimised based on the greedy algorithm, which provides personalised route customisation according to the different needs of tourists. In addition, the local travelling efficiency, the time required to visit tourist attractions and the necessary daily breaks are considered in order to reduce the cost and avoid falling into the locally optimal solution.

cs.AI↗

arXiv2024

A Personalized MOOC Learning Group and Course Recommendation Method Based on Graph Neural Network and Social Network Analysis

In order to enhance students' initiative and participation in MOOC learning, this study constructed a multi-level network model based on Social Network Analysis (SNA). The model makes use of data pertaining to nearly 40,000 users and tens of thousands of courses from various higher education MOOC platforms. Furthermore, an AI-based assistant has been developed which utilises the collected data to provide personalised recommendations regarding courses and study groups for students. The objective is to examine the relationship between students' course selection preferences and their academic interest levels. Based on the results of the relationship analysis, the AI assistant employs technologies such as GNN to recommend suitable courses and study groups to students. This study offers new insights into the potential of personalised teaching on MOOC platforms, demonstrating the value of data-driven and AI-assisted methods in improving the quality of online learning experiences, increasing student engagement, and enhancing learning outcomes.

cs.SI↗

arXiv2024

A spiral sampling method for calculating the complex orbital angular momentum spectrum

OAM spectrum reflects the OAM component included in measured light field which is crucial in OAM -based application. However, traditional definition-based OAM spectrum algorithm is extraordinary time-consuming and limited to prior knowledge severely. To overcome it, in this paper, we propose a novel method to calculate the complex spectrum. After spiral sampling and Fourier transform, one can retrieve the radial coefficient of arbitrary OAM component by proper filtering and inverse Fourier transform. The simulation results reveal that the Mean Absolute Error (MAE) between retrieved one and target could reach at 0.0221 and 0.0199 on average for amplitude and phase respectively after normalized. This method could provide a powerful tool for future OAM-based and application.

physics.optics↗

arXiv2024

If LLM Is the Wizard, Then Code Is the Wand: A Survey on How Code Empowers Large Language Models to Serve as Intelligent Agents

The prominent large language models (LLMs) of today differ from past language models not only in size, but also in the fact that they are trained on a combination of natural language and formal language (code). As a medium between humans and computers, code translates high-level goals into executable steps, featuring standard syntax, logical consistency, abstraction, and modularity. In this survey, we present an overview of the various benefits of integrating code into LLMs' training data. Specifically, beyond enhancing LLMs in code generation, we observe that these unique properties of code help (i) unlock the reasoning ability of LLMs, enabling their applications to a range of more complex natural language tasks; (ii) steer LLMs to produce structured and precise intermediate steps, which can then be connected to external execution ends through function calls; and (iii) take advantage of code compilation and execution environment, which also provides diverse feedback for model improvement. In addition, we trace how these profound capabilities of LLMs, brought by code, have led to their emergence as intelligent agents (IAs) in situations where the ability to understand instructions, decompose goals, plan and execute actions, and refine from feedback are crucial to their success on downstream tasks. Finally, we present several key challenges and future directions of empowering LLMs with code.

cs.CL↗