Alpha mining, a critical component in quantitative investment, focuses on discovering predictive signals for future asset returns in increasingly complex financial markets. However, the pervasive issue of alpha decay-where factors lose their predictive power over time-poses a significant challenge for alpha mining. Traditional methods such as genetic programming are prone to rapid alpha decay, primarily due to their susceptibility to overfitting. At the same time, approaches driven by Large Language Models (LLMs), despite their promise, often fail to impose regularization against factor homogenization-resulting in crowded signals and accelerated decay. To address this challenge, we propose AlphaAgent, an autonomous framework that effectively integrates LLM-driven agents with ad hoc regularization for mining decay-resistant alpha factors. AlphaAgent employs three key mechanisms: (i) originality enforcement through a similarity measure based on abstract syntax trees (ASTs) against existing alphas(ii) hypothesis-factor alignment via LLM-evaluated semantic consistency between market hypotheses and generated factors, and (iii) complexity control via AST-based structural constraints, preventing over-engineered constructions that are prone to overfitting. These mechanisms collectively guide the alpha generation process to balance originality, financial rationale, and adaptability to evolving market conditions, mitigating the risk of alpha decay. Extensive evaluations show that AlphaAgent outperforms traditional and LLM-based methods in mitigating alpha decay across bull and bear markets, consistently delivering significant alpha in Chinese CSI 500 and U.S. S&P 500 markets over the past four years. Notably, AlphaAgent showcases remarkable resistance to alpha decay, elevating the potential for yielding powerful factors.

BACKGROUND: Single-cell RNA sequencing (scRNA-seq) is the most widely used technique to obtain gene expression profiles from complex tissues. Cell subsets and developmental states are often identified via differential gene expression patterns. Most of the single-cell tools utilized highly variable genes to annotate cell subsets and states. However, we have discovered that a group of genes, which sensitively respond to environmental stimuli with high coefficients of variation (CV), might impose overwhelming influences on the cell type annotation. RESULT: In this research, we developed a method, based on the CV-rank and Shannon entropy, to identify these noise genes, and termed them as "sensitive genes". To validate the reliability of our methods, we applied our tools in 11 single-cell data sets from different human tissues. The results showed that most of the sensitive genes were enriched pathways related to cellular stress response. Furthermore, we noticed that the unsupervised result was closer to the ground-truth cell labels, after removing the sensitive genes detected by our tools. CONCLUSION: Our study revealed the prevalence of stochastic gene expression patterns in most types of cells, compared the differences among cell marker genes, housekeeping genes (HK genes), and sensitive genes, demonstrated the similarities of functions of sensitive genes in various scRNA-seq data sets, and improved the results of unsupervised clustering towards the ground-truth labels. We hope our method would provide new insights into the reduction of data noise in scRNA-seq data analysis and contribute to the development of better scRNA-seq unsupervised clustering algorithms in the future.

Air quality affects social activities and human health. Air quality prediction, especially for extreme events such as severe haze pollution, plays an essential guiding role in government decision-making and outdoor activity scheduling. Established prediction models face the challenges of forecasting extreme values and long-term tendency. In this paper, we propose an extreme value attention network (EvaNet) based on encoder and decoder framework to achieve long-term air quality prediction. This model designs an extreme value attention mechanism to alleviate the impact of sudden changes on prediction. In addition, to capture long-term dependence relationships, EvaNet introduces a temporal attention mechanism. Integrating the dual attention mechanisms, the extracted features are fed into a decoder to yield the final prediction. The experiments evaluated on two real-world air quality datasets show the superiority of our method against other state-of-the-art baselines.