arXiv subjects

Qiang Guan

Publications and source records attributed to Qiang Guan.

At least 37 records · Page 2Linked to original sources

arXiv2026

Readout-Side Bypass for Residual Hybrid Quantum-Classical Models

Quantum machine learning (QML) promises compact and expressive representations, but suffers from the measurement bottleneck - a narrow quantum-to-classical readout that limits performance and amplifies privacy risk. We propose a lightweight residual hybrid architecture that concatenates quantum features with raw inputs before classification, bypassing the bottleneck without increasing quantum complexity. Experiments show our model outperforms pure quantum and prior hybrid models in both centralized and federated settings. It achieves up to +55% accuracy improvement over quantum baselines, while retaining low communication cost and enhanced privacy robustness. Ablation studies confirm the effectiveness of the residual connection at the quantum-classical interface. Our method offers a practical, near-term pathway for integrating quantum models into privacy-sensitive, resource-constrained settings like federated edge learning.

cs.CR↗

arXiv2026

TokenSeek: Memory Efficient Fine Tuning via Instance-Aware Token Ditching

Fine tuning has been regarded as a de facto approach for adapting large language models (LLMs) to downstream tasks, but the high training memory consumption inherited from LLMs makes this process inefficient. Among existing memory efficient approaches, activation-related optimization has proven particularly effective, as activations consistently dominate overall memory consumption. Although prior arts offer various activation optimization strategies, their data-agnostic nature ultimately results in ineffective and unstable fine tuning. In this paper, we propose TokenSeek, a universal plugin solution for various transformer-based models through instance-aware token seeking and ditching, achieving significant fine-tuning memory savings (e.g., requiring only 14.8% of the memory on Llama3.2 1B) with on-par or even better performance. Furthermore, our interpretable token seeking process reveals the underlying reasons for its effectiveness, offering valuable insights for future research on token efficiency. Homepage: https://runjia.tech/iclr_tokenseek/

cs.CL↗

arXiv2025

Towards Explainable Quantum AI: Informing the Encoder Selection of Quantum Neural Networks via Visualization

Quantum Neural Networks (QNNs) represent a promising fusion of quantum computing and neural network architectures, offering speed-ups and efficient processing of high-dimensional, entangled data. A crucial component of QNNs is the encoder, which maps classical input data into quantum states. However, choosing suitable encoders remains a significant challenge, largely due to the lack of systematic guidance and the trial-and-error nature of current approaches. This process is further impeded by two key challenges: (1) the difficulty in evaluating encoded quantum states prior to training, and (2) the lack of intuitive methods for analyzing an encoder's ability to effectively distinguish data features. To address these issues, we introduce a novel visualization tool, XQAI-Eyes, which enables QNN developers to compare classical data features with their corresponding encoded quantum states and to examine the mixed quantum states across different classes. By bridging classical and quantum perspectives, XQAI-Eyes facilitates a deeper understanding of how encoders influence QNN performance. Evaluations across diverse datasets and encoder designs demonstrate XQAI-Eyes's potential to support the exploration of the relationship between encoder design and QNN effectiveness, offering a holistic and transparent approach to optimizing quantum encoders. Moreover, domain experts used XQAI-Eyes to derive two key practices for quantum encoder selection, grounded in the principles of pattern preservation and feature mapping.

quant-ph↗

arXiv2025

EVLP:Learning Unified Embodied Vision-Language Planner with Reinforced Supervised Fine-Tuning

In complex embodied long-horizon manipulation tasks, effective task decomposition and execution require synergistic integration of textual logical reasoning and visual-spatial imagination to ensure efficient and accurate operation. Current methods fail to adopt a unified generation framework for multimodal planning, lead to inconsistent in multimodal planning. To address this challenge, we present \textbf{EVLP (Embodied Vision-Language Planner)}, an innovative multimodal unified generation framework that jointly models linguistic reasoning and visual generation. Our approach achieves multimodal planning for long-horizon tasks through a novel training pipeline incorporating dynamic pretraining and reinforced alignment. Our core innovations consist of three key components: \textbf{1) Unified Multimodal Generation Framework}: For understanding, We integrate semantic information with spatial features to provide comprehensive visual perception. For generation, we directly learn the joint distribution of discrete images for one-step visual synthesis, enabling coordinated language-visual modeling through learnable cross-modal attention mechanisms. \textbf{2) Dynamic Perception Pretraining}: We propose a bidirectional dynamic alignment strategy employing inverse dynamics tasks and forward dynamics tasks, effectively strengthening multimodal correlations within a unified feature space. \textbf{3) Reinforced Supervised Fine-Tuning}: While conducting instruction-based fine-tuning in the unified generation space, we construct a reinforce loss to align the spatial logic between textual actions and generated images, enabling the model to acquire spatio-awared multimodal planning capabilities.

cs.CV↗

arXiv2025

A Hybrid Quantum-AI Framework for Protein Structure Prediction on NISQ Devices

Variational quantum algorithms provide a direct, physics-based approach to protein structure prediction, but their accuracy is limited by the coarse resolution of the energy landscapes generated on current noisy devices. We propose a hybrid framework that combines quantum computation with deep learning, formulating structure prediction as a problem of energy fusion. Candidate conformations are obtained through the Variational Quantum Eigensolver (VQE) executed on IBM's 127-qubit superconducting processor, which defines a global yet low-resolution quantum energy surface. To refine these basins, secondary structure probabilities and dihedral angle distributions predicted by the NSP3 neural network are incorporated as statistical potentials. These additional terms sharpen the valleys of the quantum landscape, resulting in a fused energy function that enhances effective resolution and better distinguishes native-like structures. Evaluation on 375 conformations from 75 protein fragments shows consistent improvements over AlphaFold3, ColabFold, and quantum-only predictions, achieving a mean RMSD of 4.9 Å with statistical significance (p < 0.001). The findings demonstrate that energy fusion offers a systematic method for combining data-driven models with quantum algorithms, improving the practical applicability of near-term quantum computing to molecular and structural biology.

cs.ET↗

arXiv2025

KIS-S: A GPU-Aware Kubernetes Inference Simulator with RL-Based Auto-Scaling

Autoscaling GPU inference workloads in Kubernetes remains challenging due to the reactive and threshold-based nature of default mechanisms such as the Horizontal Pod Autoscaler (HPA), which struggle under dynamic and bursty traffic patterns and lack integration with GPU-level metrics. We present KIS-S, a unified framework that combines KISim, a GPU-aware Kubernetes Inference Simulator, with KIScaler, a Proximal Policy Optimization (PPO)-based autoscaler. KIScaler learns latency-aware and resource-efficient scaling policies entirely in simulation, and is directly deployed without retraining. Experiments across four traffic patterns show that KIScaler improves average reward by 75.2%, reduces P95 latency up to 6.7x over CPU baselines, and generalizes without retraining. Our work bridges the gap between reactive autoscaling and intelligent orchestration for scalable GPU-accelerated environments.

cs.DC↗

arXiv2025

QDockBank: A Dataset for Ligand Docking on Protein Fragments Predicted on Utility-Level Quantum Computers

Protein structure prediction is a core challenge in computational biology, particularly for fragments within ligand-binding regions, where accurate modeling is still difficult. Quantum computing offers a novel first-principles modeling paradigm, but its application is currently limited by hardware constraints, high computational cost, and the lack of a standardized benchmarking dataset. In this work, we present QDockBank-the first large-scale protein fragment structure dataset generated entirely using utility-level quantum computers, specifically designed for protein-ligand docking tasks. QDockBank comprises 55 protein fragments extracted from ligand-binding pockets. The dataset was generated through tens of hours of execution on superconducting quantum processors, making it the first quantum-based protein structure dataset with a total computational cost exceeding one million USD. Experimental evaluations demonstrate that structures predicted by QDockBank outperform those predicted by AlphaFold2 and AlphaFold3 in terms of both RMSD and docking affinity scores. QDockBank serves as a new benchmark for evaluating quantum-based protein structure prediction.

cs.ET↗

arXiv2025

Adaptive Job Scheduling in Quantum Clouds Using Reinforcement Learning

Present-day quantum systems face critical bottlenecks, including limited qubit counts, brief coherence intervals, and high susceptibility to errors-all of which obstruct the execution of large and complex circuits. The advancement of quantum algorithms has outpaced the capabilities of existing quantum hardware, making it difficult to scale computations effectively. Additionally, inconsistencies in hardware performance and pervasive quantum noise undermine system stability and computational accuracy. To optimize quantum workloads under these constraints, strategic approaches to task scheduling and resource coordination are essential. These methods must aim to accelerate processing, retain operational fidelity, and reduce the communication burden inherent to distributed setups. One of the persistent challenges in this domain is how to efficiently divide and execute large circuits across multiple quantum processors (QPUs), especially in error-prone environments. In response, we introduce a simulation-based tool that supports distributed scheduling and concurrent execution of quantum jobs on networked QPUs connected via real-time classical channels. The tool models circuit decomposition for workloads that surpass individual QPU limits, allowing for parallel execution through inter-processor communication. Using this simulation environment, we compare four distinct scheduling techniques-among them, a model informed by reinforcement learning. These strategies are evaluated across multiple metrics, including runtime efficiency, fidelity preservation, and communication costs. Our analysis underscores the trade-offs inherent in each approach and highlights how parallelized, noise-aware scheduling can meaningfully improve computational throughput in distributed quantum infrastructures.

cs.DC↗

arXiv2025

Can Large Language Models Understand Intermediate Representations in Compilers?

Intermediate Representations (IRs) play a critical role in compiler design and program analysis, yet their comprehension by Large Language Models (LLMs) remains underexplored. In this paper, we present an explorative empirical study evaluating the capabilities of six state-of-the-art LLMs: GPT-4, GPT-3, DeepSeek, Gemma 2, Llama 3, and Code Llama, in understanding IRs. Specifically, we assess model performance across four core tasks: control flow graph reconstruction, decompilation, code summarization, and execution reasoning. While LLMs exhibit competence in parsing IR syntax and identifying high-level structures, they consistently struggle with instruction-level reasoning, especially in control flow reasoning, loop handling, and dynamic execution. Common failure modes include misinterpreting branching instructions, omitting critical operations, and relying on heuristic reasoning rather than precise instruction-level logic. Our findings highlight the need for IR-specific enhancements in LLM design. We recommend fine-tuning on structured IR datasets and integrating control-flow-sensitive architectures to improve model effectiveness. All experimental data and source code are publicly available at

cs.LG↗

arXiv2025

Intuit: Explain Quantum Computing Concepts via AR-based Analogy

Quantum computing has shown great potential to revolutionize traditional computing and can provide an exponential speedup for a wide range of possible applications, attracting various stakeholders. However, understanding fundamental quantum computing concepts remains a significant challenge for novices because of their abstract and counterintuitive nature. Thus, we propose an analogy-based characterization framework to construct the mental mapping between quantum computing concepts and daily objects, informed by in-depth expert interviews and a literature review, covering key quantum concepts and characteristics like number of qubits, output state duality, quantum concept type, and probability quantification. Then, we developed an AR-based prototype system, Intuit, using situated analytics to explain quantum concepts through daily objects and phenomena (e.g., rotating coins, paper cutters). We thoroughly evaluated our approach through in-depth user and expert interviews. The Results demonstrate the effectiveness and usability of Intuit in helping learners understand abstract concepts in an intuitive and engaging manner.

cs.HC↗

arXiv2025

Escaping Barren Plateau: Co-Exploration of Quantum Circuit Parameters and Architectures

Barren plateaus (BP), characterized by exponentially vanishing gradients that hinder the training of variational quantum circuits (VQC), present a pervasive and critical challenge in applying variational quantum algorithms to real-world applications. It is widely recognized that the BP problem becomes more pronounced with an increase in the number of parameters. This work demonstrates that the BP problem manifests at different scales depending on the specific application, highlighting the absence of a universal VQC ansatz capable of resolving the BP issue across all applications. Consequently, there is an imminent need for an automated tool to design and optimize VQC architectures tailored to specific applications. To close the gap, this paper takes Variational Quantum Eigensolvers (VQEs) as a vehicle, and we propose a novel quantum circuit parameter and architecture co-exploration framework, namely AntiBP. Experimental results demonstrate that AntiBP effectively avoids the BP issue for circuits that are not under-parameterized in noise-free environments. Furthermore, AntiBP significantly outperforms baseline VQEs in noisy environments.

quant-ph↗

arXiv2024

Adopting Trustworthy AI for Sleep Disorder Prediction: Deep Time Series Analysis with Temporal Attention Mechanism and Counterfactual Explanations

Sleep disorders have a major impact on both lifestyle and health. Effective sleep disorder prediction from lifestyle and physiological data can provide essential details for early intervention. This research utilizes three deep time series models and facilitates them with explainability approaches for sleep disorder prediction. Specifically, our approach adopts Temporal Convolutional Networks (TCN), Long Short-Term Memory (LSTM) for time series data analysis, and Temporal Fusion Transformer model (TFT). Meanwhile, the temporal attention mechanism and counterfactual explanation with SHapley Additive exPlanations (SHAP) approach are employed to ensure dependable, accurate, and interpretable predictions. Finally, using a large dataset of sleep health measures, our evaluation demonstrates the effect of our method in predicting sleep disorders.

cs.LG↗

arXiv2024

MNIST-Fraction: Enhancing Math Education with AI-Driven Fraction Detection and Analysis

Mathematics education, a crucial and basic field, significantly influences students' learning in related subjects and their future careers. Utilizing artificial intelligence to interpret and comprehend math problems in education is not yet fully explored. This is due to the scarcity of quality datasets and the intricacies of processing handwritten information. In this paper, we present a novel contribution to the field of mathematics education through the development of MNIST-Fraction, a dataset inspired by the renowned MNIST, specifically tailored for the recognition and understanding of handwritten math fractions. Our approach is the utilization of deep learning, specifically Convolutional Neural Networks (CNNs), for the recognition and understanding of handwritten math fractions to effectively detect and analyze fractions, along with their numerators and denominators. This capability is pivotal in calculating the value of fractions, a fundamental aspect of math learning. The MNIST-Fraction dataset is designed to closely mimic real-world scenarios, providing a reliable and relevant resource for AI-driven educational tools. Furthermore, we conduct a comprehensive comparison of our dataset with the original MNIST dataset using various classifiers, demonstrating the effectiveness and versatility of MNIST-Fraction in both detection and classification tasks. This comparative analysis not only validates the practical utility of our dataset but also offers insights into its potential applications in math education. To foster collaboration and further research within the computational and educational communities. Our work aims to bridge the gap in high-quality educational resources for math learning, offering a valuable tool for both educators and researchers in the field.

cs.LG↗

arXiv2024

Approaches to Simultaneously Solving Variational Quantum Eigensolver Problems

The variational quantum eigensolver (VQE), a type of variational quantum algorithm, is a hybrid quantum-classical algorithm to find the lowest-energy eigenstate of a particular Hamiltonian. We investigate ways to optimize the VQE solving process on multiple instances of the same problem, by observing the process on one instance of the problem to inform initialization for other processes. We aim to take advantage of the VQE solution process to obtain useful information while disregarding information which we can predict to not be very useful. In particular, we find that the solution process produces lots of data with very little new information. Therefore, we can safely disregard much of this repetitive information with little effect on the outcome of the solution process.

quant-ph↗

arXiv2024

Efficient Circuit Wire Cutting Based on Commuting Groups

Current quantum devices face challenges when dealing with large circuits due to error rates as circuit size and the number of qubits increase. The circuit wire-cutting technique addresses this issue by breaking down a large circuit into smaller, more manageable subcircuits. However, the exponential increase in the number of subcircuits and the complexity of reconstruction as more cuts are made poses a great practical challenge. Inspired by ancilla-assisted quantum process tomography and the MUBs-based grouping technique for simultaneous measurement, we propose a new approach that can reduce subcircuit running overhead. The approach first uses ancillary qubits to transform all quantum input initializations into quantum output measurements. These output measurements are then organized into commuting groups for the purpose of simultaneous measurement, based on MUBs-based grouping. This approach significantly reduces the number of necessary subcircuits as well as the total number of shots. Lastly, we provide numerical experiments to demonstrate the complexity reduction.

quant-ph↗

arXiv2024

Radiance Field Learners As UAV First-Person Viewers

First-Person-View (FPV) holds immense potential for revolutionizing the trajectory of Unmanned Aerial Vehicles (UAVs), offering an exhilarating avenue for navigating complex building structures. Yet, traditional Neural Radiance Field (NeRF) methods face challenges such as sampling single points per iteration and requiring an extensive array of views for supervision. UAV videos exacerbate these issues with limited viewpoints and significant spatial scale variations, resulting in inadequate detail rendering across diverse scales. In response, we introduce FPV-NeRF, addressing these challenges through three key facets: (1) Temporal consistency. Leveraging spatio-temporal continuity ensures seamless coherence between frames; (2) Global structure. Incorporating various global features during point sampling preserves space integrity; (3) Local granularity. Employing a comprehensive framework and multi-resolution supervision for multi-scale scene feature representation tackles the intricacies of UAV video spatial scales. Additionally, due to the scarcity of publicly available FPV videos, we introduce an innovative view synthesis method using NeRF to generate FPV perspectives from UAV footage, enhancing spatial perception for drones. Our novel dataset spans diverse trajectories, from outdoor to indoor environments, in the UAV domain, differing significantly from traditional NeRF scenarios. Through extensive experiments encompassing both interior and exterior building structures, FPV-NeRF demonstrates a superior understanding of the UAV flying space, outperforming state-of-the-art methods in our curated UAV dataset. Explore our project page for further insights: https://fpv-nerf.github.io/.

cs.CV↗

arXiv2024

AMD: Automatic Multi-step Distillation of Large-scale Vision Models

Transformer-based architectures have become the de-facto standard models for diverse vision tasks owing to their superior performance. As the size of the models continues to scale up, model distillation becomes extremely important in various real applications, particularly on devices limited by computational resources. However, prevailing knowledge distillation methods exhibit diminished efficacy when confronted with a large capacity gap between the teacher and the student, e.g, 10x compression rate. In this paper, we present a novel approach named Automatic Multi-step Distillation (AMD) for large-scale vision model compression. In particular, our distillation process unfolds across multiple steps. Initially, the teacher undergoes distillation to form an intermediate teacher-assistant model, which is subsequently distilled further to the student. An efficient and effective optimization framework is introduced to automatically identify the optimal teacher-assistant that leads to the maximal student performance. We conduct extensive experiments on multiple image classification datasets, including CIFAR-10, CIFAR-100, and ImageNet. The findings consistently reveal that our approach outperforms several established baselines, paving a path for future knowledge distillation methods on large-scale vision models.

cs.CV↗

arXiv2024

Which One Changes More? A Novel Radial Visualization for State Change Comparison

It is common to compare state changes of multiple data items and identify which data items have changed more in various applications (e.g., annual GDP growth of different countries and daily increase of new COVID-19 cases in different regions). Grouped bar charts and slope graphs can visualize both state changes and their initial and final states of multiple data items, and are thus widely used for state change comparison. But they leverage implicit bar differences or line slopes to indicate state changes, which has been proven less effective for visual comparison. Both visualizations also suffer from visual scalability issues when an increasing number of data items need to be compared. This paper fills the research gap by proposing a novel radial visualization called Intercept Graph to facilitate visual comparison of multiple state changes. It consists of inner and outer axes, and leverages the lengths of line segments intercepted by the inner axis to explicitly encode the state changes. Users can interactively adjust the inner axis to filter large changes of their interest and magnify the difference of relatively-similar state changes, enhancing its visual scalability and comparison accuracy. We extensively evaluate the Intercept Graph in comparison with baseline methods through two usage scenarios, quantitative metric evaluations, and well-designed crowdsourcing user studies with 50 participants. Our results demonstrate the usefulness and effectiveness of the Intercept Graph.

cs.HC↗