Quality-Driven Agentic Reasoning for LLM-Assisted Software Design: Questions-of-Thoughts (QoT) as a Time-Series Self-QA Chain

Recent advances in large language models (LLMs) have accelerated AI-assisted software development, yet practical deployment remains constrained by incomplete implementations, weak modularization, and inconsistent security practices. We introduce Questions-of-Thoughts (QoT), a quality-driven inference-time scaffold that turns a user goal into (i) an ordered sequence of engineering steps and (ii) stepwise self-questioning to verify constraints and reduce omission errors, while maintaining a lightweight reasoning record that stabilizes subsequent design decisions. We evaluate QoT across three representative backend engineering domains: API Design, Data Communication, and File Systems. Each task requires multi-module decomposition and exposes standard failure modes in LLM-generated systems. To enable data-driven comparison, we score generated artifacts using an ISO/IEC-inspired quality rubric that measures Scalability, Completeness, Modularity, and Security. We report domain-wise gains as the change in total quality score, defined as the QoT score minus the NoQoT score. Results show capacity-dependent improvements: QoT yields consistent quality improvements for larger models and more complex domains, while smaller models may exhibit trade-offs under tight context and planning budgets. We release an open artifact with prompts, scoring guidelines, raw generations, and scripts that reproduce the reported tables and figures to support applied AI and data analytics research.

Quantum Machine Learning, Quantitative Trading, Reinforcement Learning, Deep Learning

The convergence of quantum-inspired neural networks and deep reinforcement learning offers a promising avenue for financial trading. We implemented a trading agent for USD/TWD by integrating Quantum Long Short-Term Memory (QLSTM) for short-term trend prediction with Quantum Asynchronous Advantage Actor-Critic (QA3C), a quantum-enhanced variant of the classical A3C. Trained on data from 2000-01-01 to 2025-04-30 (80\% training, 20\% testing), the long-only agent achieves 11.87\% return over around 5 years with 0.92\% max drawdown, outperforming several currency ETFs. We detail state design (QLSTM features and indicators), reward function for trend-following/risk control, and multi-core training. Results show hybrid models yield competitive FX trading performance. Implications include QLSTM's effectiveness for small-profit trades with tight risk and future enhancements. Key hyperparameters: QLSTM sequence length$=$4, QA3C workers$=$8. Limitations: classical quantum simulation and simplified strategy. \footnote{The views expressed in this article are those of the authors and do not represent the views of Wells Fargo. This article is for informational purposes only. Nothing contained in this article should be construed as investment advice. Wells Fargo makes no express or implied warranties and expressly disclaims all legal, tax, and accounting implications related to this article.

Benchmarking Quantum and Classical Sequential Models for Urban Telecommunication Forecasting

In this study, we evaluate the performance of classical and quantum-inspired sequential models in forecasting univariate time series of incoming SMS activity (SMS-in) using the Milan Telecommunication Activity Dataset. Due to data completeness limitations, we focus exclusively on the SMS-in signal for each spatial grid cell. We compare five models, LSTM (baseline), Quantum LSTM (QLSTM), Quantum Adaptive Self-Attention (QASA), Quantum Receptance Weighted Key-Value (QRWKV), and Quantum Fast Weight Programmers (QFWP), under varying input sequence lengths (4, 8, 12, 16, 32 and 64). All models are trained to predict the next 10-minute SMS-in value based solely on historical values within a given sequence window. Our findings indicate that different models exhibit varying sensitivities to sequence length, suggesting that quantum enhancements are not universally advantageous. Rather, the effectiveness of quantum modules is highly dependent on the specific task and architectural design, reflecting inherent trade-offs among model size, parameterization strategies, and temporal modeling capabilities.

Robust and Annotation-Free Wound Segmentation on Noisy Real-World Pressure Ulcer Images: Towards Automated DESIGN-R\textsuperscript{\textregistered} Assessment

Purpose: Accurate wound segmentation is essential for automated DESIGN-R scoring. However, existing models such as FUSegNet, which are trained primarily on foot ulcer datasets, often fail to generalize to wounds on other body sites. Methods: We propose an annotation-efficient pipeline that combines a lightweight YOLOv11n-based detector with the pre-trained FUSegNet segmentation model. Instead of relying on pixel-level annotations or retraining for new anatomical regions, our method achieves robust performance using only 500 manually labeled bounding boxes. This zero fine-tuning approach effectively bridges the domain gap and enables direct deployment across diverse wound types. This is an advance not previously demonstrated in the wound segmentation literature. Results: Evaluated on three real-world test sets spanning foot, sacral, and trochanter wounds, our YOLO plus FUSegNet pipeline improved mean IoU by 23 percentage points over vanilla FUSegNet and increased end-to-end DESIGN-R size estimation accuracy from 71 percent to 94 percent (see Table 3 for details). Conclusion: Our pipeline generalizes effectively across body sites without task-specific fine-tuning, demonstrating that minimal supervision, with 500 annotated ROIs, is sufficient for scalable, annotation-light wound segmentation. This capability paves the way for real-world DESIGN-R automation, reducing reliance on pixel-wise labeling, streamlining documentation workflows, and supporting objective and consistent wound scoring in clinical practice. We will publicly release the trained detector weights and configuration to promote reproducibility and facilitate downstream deployment.

Quantum Feature Optimization for Enhanced Clustering of Blockchain Transaction Data

Blockchain transaction data exhibits high dimensionality, noise, and intricate feature entanglement, presenting significant challenges for traditional clustering algorithms. In this study, we conduct a comparative analysis of three clustering approaches: (1) Classical K-Means Clustering, applied to pre-processed feature representations; (2) Hybrid Clustering, wherein classical features are enhanced with quantum random features extracted using randomly initialized quantum neural networks (QNNs); and (3) Fully Quantum Clustering, where a QNN is trained in a self-supervised manner leveraging a SwAV-based loss function to optimize the feature space for clustering directly. The proposed experimental framework systematically investigates the impact of quantum circuit depth and the number of learned prototypes, demonstrating that even shallow quantum circuits can effectively extract meaningful non-linear representations, significantly improving clustering performance.

Financial Vision Based Reinforcement Learning Trading Strategy

Recent advances in artificial intelligence (AI) for quantitative trading have led to its general superhuman performance in significant trading performance. However, the potential risk of AI trading is a "black box" decision. Some AI computing mechanisms are complex and challenging to understand. If we use AI without proper supervision, AI may lead to wrong choices and make huge losses. Hence, we need to ask about the AI "black box", including why did AI decide to do this or not? Why can people trust AI or not? How can people fix their mistakes? These problems also highlight the challenges that AI technology can explain in the trading field.

Dynamic Deep Convolutional Candlestick Learner

Candlestick pattern is one of the most fundamental and valuable graphical tools in financial trading that supports traders observing the current market conditions to make the proper decision. This task has a long history and, most of the time, human experts. Recently, efforts have been made to automatically classify these patterns with the deep learning models. The GAF-CNN model is a well-suited way to imitate how human traders capture the candlestick pattern by integrating spatial features visually. However, with the great potential of the GAF encoding, this classification task can be extended to a more complicated object detection level. This work presents an innovative integration of modern object detection techniques and GAF time-series encoding on candlestick pattern tasks. We make crucial modifications to the representative yet straightforward YOLO version 1 model based on our time-series encoding method and the property of such data type. Powered by the deep neural networks and the unique architectural design, the proposed model performs pretty well in candlestick classification and location recognition. The results show tremendous potential in applying modern object detection techniques on time-series tasks in a real-time manner.

Financial Vision Based Differential Privacy Applications

The importance of deep learning data privacy has gained significant attention in recent years. It is probably to suffer data breaches when applying deep learning to cryptocurrency that lacks supervision of financial regulatory agencies. However, there is little relative research in the financial area to our best knowledge. We apply two representative deep learning privacy-privacy frameworks proposed by Google to financial trading data. We designed the experiments with several different parameters suggested from the original studies. In addition, we refer the degree of privacy to Google and Apple companies to estimate the results more reasonably. The results show that DP-SGD performs better than the PATE framework in financial trading data. The tradeoff between privacy and accuracy is low in DP-SGD. The degree of privacy also is in line with the actual case. Therefore, we can obtain a strong privacy guarantee with precision to avoid potential financial loss.

Data Augmentation for Deep Candlestick Learner

To successfully build a deep learning model, it will need a large amount of labeled data. However, labeled data are hard to collect in many use cases. To tackle this problem, a bunch of data augmentation methods have been introduced recently and have demonstrated successful results in computer vision, natural language and so on. For financial trading data, to our best knowledge, successful data augmentation framework has rarely been studied. Here we propose a Modified Local Search Attack Sampling method to augment the candlestick data, which is a very important tool for professional trader. Our results show that the proposed method can generate high-quality data which are hard to distinguish by human and will open a new way for finance community to employ existing machine learning techniques even if the dataset is small.

Adversarial Robustness of Deep Convolutional Candlestick Learner

Deep learning (DL) has been applied extensively in a wide range of fields. However, it has been shown that DL models are susceptible to a certain kinds of perturbations called \emph{adversarial attacks}. To fully unlock the power of DL in critical fields such as financial trading, it is necessary to address such issues. In this paper, we present a method of constructing perturbed examples and use these examples to boost the robustness of the model. Our algorithm increases the stability of DL models for candlestick classification with respect to perturbations in the input data.