Paper: arXiv 2509.10461

Authors: Hao Wang, Jingshu Peng, Yanyan Shen, Xujia Li, Lei Chen

Abstract

Stock recommendation is critical in Fintech applications, which use price series and alternative information to estimate future stock performance. Although deep learning models are prevalent in stock recommendation systems, traditional time-series forecasting training often fails to capture stock trends and rankings simultaneously, which are essential consideration factors for investors. To tackle this issue, we introduce a Multi-Task Learning (MTL) framework for stock recommendation, \textbf{“M”}omentum-\textbf{“i”}ntegrated \textbf{“M”}ulti-task \textbf{“Stoc”}k \textbf{“R”}ecommendation with Converge-based Optimization (\textbf{“MiM-StocR”}). To improve the model’s ability to capture short-term trends, we novelly invoke a momentum line indicator in model training. To prioritize top-performing stocks and optimize investment allocation, we propose a list-wise ranking loss function called Adaptive-k ApproxNDCG. Moreover, due to the volatility and uncertainty of the stock market, existing MTL frameworks face overfitting issues when applied to stock time series. To mitigate this issue, we introduce the Converge-based Quad-Balancing (CQB) method. We conducted extensive experiments on three stock benchmarks: SEE50, CSI 100, and CSI 300. MiM-StocR outperforms state-of-the-art MTL baselines across both ranking and profitable evaluations.

Complexity vs Empirical Score

  • Math Complexity: 7.0/10
  • Empirical Rigor: 6.0/10 (capped from 8.5: transaction costs not modelled; see penalty flags)
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper introduces novel mathematical components including a custom Adaptive-k ApproxNDCG ranking loss function and a Converge-based Quad-Balancing (CQB) optimization method, demonstrating high technical depth. Empirically, it is rigorously tested on three real-world stock benchmarks (SEE50, CSI 100, CSI 300) with profitability evaluations, ablation studies, and open-sourced code, making it highly implementation-ready.

Research Flowchart

  flowchart TD
  A["Research Goal:<br>Improve Stock Recommendation<br>via Multi-Task Learning"] --> B["Data Inputs:<br>Price Series & Alternative Data<br>(SEE50, CSI 100, CSI 300)"]
  
  B --> C["Key Methodology:<br>MiM-StocR Framework"]
  
  C --> D["Momentum Integration:<br>Captures Short-term Trends"]
  C --> E["Adaptive-k ApproxNDCG:<br>List-wise Ranking Loss"]
  C --> F["Converge-based<br>Quad-Balancing (CQB):<br>Mitigates Overfitting"]
  
  D --> G["Multi-Task Learning Process"]
  E --> G
  F --> G
  
  G --> H["Key Findings & Outcomes:<br>1. State-of-the-art Ranking<br>2. Enhanced Profitability<br>3. Robust MTL Performance"]
  
  style A fill:#e1f5e1
  style H fill:#ffe6e6

Cited References (86)

  1. Temporal relational ranking for stock prediction (2019)
  2. DoubleAdapt: A Meta-learning Approach to Incremental Learning for Stock Trend Forecasting (2023)
  3. DeepUnifiedMom: Unified Time-series Momentum Portfolio Construction via Multi-Task Learning with Multi-Gate Mixture of Experts — → 2406.08742
  4. Pairwise ranking loss for multi-task learning in recommender systems (2024) — link
  5. Multi-Modal Temporal Dynamic Graph Construction for Stock Rank Prediction (2025)
  6. On Evaluating Loss Functions for Stock Ranking: An Empirical Analysis With Transformer Model (2025)
  7. Optimizing portfolio selection through stock ranking and matching: A reinforcement learning approach (2025)
  8. Constructing time-series momentum portfolios with deep multi-task learning — → 2306.13661
and 78 more
  1. Metabalance: improving multi-task recommendations via adapting gradient magnitudes of auxiliary tasks (2022)
  2. Momentum portfolio selection based on learning-to-rank algorithms with heterogeneous knowledge graphs (2024)
  3. Quantitative stock portfolio optimization by multi-task learning risk and return (2024)
  4. Portfolio Optimization with Multi-Trend Objective and Accelerated Quasi-Newton Method (2024)
  5. An integrated framework of deep learning and knowledge graph for prediction of stock price trend: An application in Chinese stock exchange market (2020)
  6. Deep learning for event-driven stock prediction (2015)
  7. Knowledge-driven stock trend prediction and explanation via temporal convolutional network (2019)
  8. Deep learning networks for stock market analysis and prediction: Methodology, data representations, and case studies (2017)
  9. Listening to chaotic whispers: A deep learning framework for news-oriented stock trend prediction (2018)
  10. Using a financial training criterion rather than a prediction criterion (1997)
  11. Multi-task learning as multi-objective optimization (2018)
  12. Gradient surgery for multi-task learning (2020)
  13. Stock price prediction using the ARIMA model (2014)
  14. Stock market autoregressive dynamics: a multinational comparative study with quantile regression (2016)
  15. Common risk factors in the returns on stocks and bonds (1993)
  16. Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency (1993)
  17. Expected reciprocal rank for graded relevance (2009)
  18. Stock market volatility and return analysis: A systematic literature review (2020)
  19. The Current State of Quantitative Equity Investing (2018)
  20. Factor investing: alpha concentration versus diversification (2021)
  21. Jumps in equilibrium prices and market microstructure noise (2012)
  22. Simultaneous optimization of artificial neural networks for financial forecasting (2012)
  23. Recurrent neural network and a hybrid model for prediction of stock returns (2015)
  24. Stock market forecasting using recurrent neural network (2016)
  25. A deep learning framework for financial time series using stacked autoencoders and long-short term memory (2017)
  26. Multi-task recurrent neural networks and higher-order Markov random fields for stock price movement prediction: Multi-task RNN and higer-order MRFs for stock price classification (2019)
  27. Stock price prediction via discovering multi-frequency trading patterns (2017)
  28. Enhancing stock movement prediction with adversarial training (2018)
  29. Adaptive long-short pattern transformer for stock investment selection (2022)
  30. A theoretical analysis of NDCG type ranking measures (2013)
  31. Stock selection via spatiotemporal hypergraph attention network: A learning to rank approach (2021)
  32. Hierarchical Multi-Scale Gaussian Transformer for Stock Movement Prediction. (2020)
  33. Hist: A graph-based framework for stock trend forecasting via mining concept-oriented shared information (2021)
  34. Knowledge graph-based event embedding framework for financial quantitative investments (2020)
  35. Exploring graph neural networks for stock market predictions with rolling window analysis (2019)
  36. Qlib: An ai-oriented quantitative investment platform (2020)
  37. Methods for Acquiring and Incorporating Knowledge into Stock Price Prediction: A Survey (2023)
  38. Revisiting approximate metric optimization in the age of deep neural networks (2019)
  39. Graph-based stock recommendation by time-aware relational attention network (2021)
  40. Learning multiple stock trading patterns with temporal routing adaptor and optimal transport (2021)
  41. Multitask learning: A knowledge-based source of inductive bias1 (1993)
  42. Reversible instance normalization for accurate time-series forecasting against distribution shift (2021)
  43. Which tasks should be learned together in multi-task learning? (2020)
  44. Multi-task learning using uncertainty to weigh losses for scene geometry and semantics (2018)
  45. A survey on multi-task learning (2021)
  46. Gradnorm: Gradient normalization for adaptive loss balancing in deep multitask networks (2018)
  47. Conflict-averse gradient descent for multi-task learning (2021)
  48. Long short-term memory (1997)
  49. Graph attention networks (2017)
  50. FAMO: Fast Adaptive Multitask Optimization (2023)
  51. Towards impartial multi-task learning (2021)
  52. Multi-task learning as a bargaining game (2022)
  53. Multi-task learning for stock selection (1996)
  54. Stock ranking with multi-task learning (2022)
  55. Stock market forecasting using a multi-task approach integrating long short-term memory and the random forest framework (2022)
  56. Html: Hierarchical transformer-based multi-task learning for volatility prediction (2020)
  57. Multi-task Envisioning Transformer-based Autoencoder for Corporate Credit Rating Migration Early Prediction (2022)
  58. NumHTML: Numeric-Oriented Hierarchical Transformer Model for Multi-task Financial Forecasting (2022)
  59. Day trading profit maximization with multi-task learning and technical analysis (2015)
  60. Fact, fiction, and momentum investing (2014)
  61. Momentum has its moments (2015)
  62. An overview of multi-task learning in deep neural networks (2017)
  63. A general approximation framework for direct optimization of information retrieval measures (2010)
  64. Two-person cooperative games (1953)
  65. Learning phrase representations using RNN encoder-decoder for statistical machine translation (2014)
  66. A Time Series is Worth 64 Words: Long-term Forecasting with Transformers (2022)
  67. The worst, the best, ignoring all the rest: The rank effect and trading behavior (2015)
  68. Dual-Balancing for Multi-Task Learning (2023)
  69. Independent Component Alignment for Multi-Task Learning (2023)
  70. Indoor segmentation and support inference from rgbd images (2012)
  71. The cityscapes dataset for semantic urban scene understanding (2016)
  72. Depth map prediction from a single image using a multi-scale deep network (2014)
  73. Reasonable effectiveness of random weighting: A litmus test for multi-task learningsawhney2021stock (2021)
  74. Stock ranking prediction using list-wise approach and node embedding technique (2021)
  75. Learning to rank: from pairwise approach to listwise approach (2007)
  76. Constructing long-short stock portfolio with a new listwise learn-to-rank algorithm (2022)
  77. Stock market prediction using machine learning techniques: a decade survey on methodologies, recent developments, and future directions (2021)
  78. Stock Market Prediction via Deep Learning Techniques: A Survey (2022)

2 of 86 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.