Vector-Quantized Discrete Latent Factors Meet Financial Priors: Dynamic Cross-Sectional Stock Ranking Prediction for Portfolio Construction

• URL: http://arxiv.org/abs/2605.13407v1
• Date: Wed, 13 May 2026 12:02:53 GMT
• Title: Vector-Quantized Discrete Latent Factors Meet Financial Priors: Dynamic Cross-Sectional Stock Ranking Prediction for Portfolio Construction
• Abstract summary: PRISM-VQ (PRior-Informed Stock Model with Vector Quantization) is a dynamic factor framework that integrates expert prior factors, vector-quantized latent factors learned from cross-sectional structure, and a structure-conditioned Mixture-of-Experts to generate time-varying factor loadings.<n> Experiments on CSI 300 and S&P 500 show consistent improvements in cross-sectional return prediction and portfolio performance over strong baselines while preserving interpretability.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Predicting cross-sectional stock returns is challenging due to low signal-to-noise ratios and evolving market regimes. Classical factor models offer interpretability but limited flexibility, while deep learning models achieve strong performance yet often underutilize financial priors. We address this gap with PRISM-VQ (PRior-Informed Stock Model with Vector Quantization), a dynamic factor framework that integrates expert prior factors, vector-quantized discrete latent factors learned from cross-sectional structure, and a structure-conditioned Mixture-of-Experts to generate time-varying factor loadings. Vector quantization acts as an information bottleneck that suppresses noise while capturing robust market structure, with discrete codes serving both as latent factors and as routing signals for temporal expert specialization. Experiments on CSI 300 and S&P 500 show consistent improvements in cross-sectional return prediction and portfolio performance over strong baselines while preserving interpretability. Our code is available at https://github.com/finxlab/PRISM-VQ.

Related papers

• Spectral Evolution-Guided Token Pruning in Multimodal Large Language Models [46.61944018676346]
  We propose a training-free token pruning framework based on Cross-Layer Spectral Evolution (CLSE)<n>Instead of measuring token importance from single-layer feature magnitudes, CLSE quantifies how token representations evolve across Transformer layers in the frequency domain.<n>CLSE achieves a superior trade-off between efficiency and accuracy under aggressive token reduction.
  arXiv  Detail & Related papers  (2026-06-23T05:39:52Z)
• AnchorMoE: Interpretable Time Series Classification via Anchor-Routed MoE [51.994752158953084]
  AnchorMoE is an interpretable-by-construction classification framework.<n>It encodes multi-view representations of local patches and routes them to specialized experts.<n>Experiments on real-world and synthetic benchmarks demonstrate that AnchorMoE achieves highly competitive classification performance.
  arXiv  Detail & Related papers  (2026-06-02T13:30:54Z)
• MarketGANs: Multivariate financial time-series data augmentation using generative adversarial networks [4.104902299123294]
  MarketGAN is a factor-based generative framework for high-dimensional asset return generation under severe data scarcity.<n>We embed an explicit asset-pricing factor structure as an economic inductive bias and generate returns as a single joint vector.<n>Using daily returns of large U.S. equities, we find that MarketGAN more closely matches empirical stylized facts of asset returns.
  arXiv  Detail & Related papers  (2026-01-25T10:19:04Z)
• Quantum Temporal Convolutional Neural Networks for Cross-Sectional Equity Return Prediction: A Comparative Benchmark Study [8.777151099203934]
  Quantum Temporal Convolutional Neural Network (QTCNN) combines a classical temporal encoder with parameter-efficient quantum convolution circuits for cross-sectional equity return prediction.<n> QTCNN achieves a Sharpe ratio of 0.538, outperforming the best classical baseline by approximately 72%.<n>These results highlight the practical potential of quantum-enhanced forecasting model, QTCNN, for robust decision-making in quantitative finance.
  arXiv  Detail & Related papers  (2025-12-07T02:34:24Z)
• Diffolio: A Diffusion Model for Multivariate Probabilistic Financial Time-Series Forecasting and Portfolio Construction [7.0782219254786725]
  We propose Diffolio, a diffusion model designed for multivariate financial time-series forecasting and portfolio construction.<n>Diffolio employs a denoising network with a hierarchical attention architecture, comprising both asset-level and market-level layers.
  arXiv  Detail & Related papers  (2025-11-10T12:05:32Z)
• Continual Action Quality Assessment via Adaptive Manifold-Aligned Graph Regularization [53.82400605816587]
  Action Quality Assessment (AQA) quantifies human actions in videos, supporting applications in sports scoring, rehabilitation, and skill evaluation.<n>A major challenge lies in the non-stationary nature of quality distributions in real-world scenarios.<n>We introduce Continual AQA (CAQA), which equips with Continual Learning capabilities to handle evolving distributions.
  arXiv  Detail & Related papers  (2025-10-08T10:09:47Z)
• FactorGCL: A Hypergraph-Based Factor Model with Temporal Residual Contrastive Learning for Stock Returns Prediction [16.70420232738428]
  We propose a hypergraph-based factor model with temporal residual contrastive learning (FactorGCL)<n>To mine hidden factors that supplement human-designed prior factors for predicting stock returns, we design a cascading residual hypergraph architecture.<n>Our experiments on real stock market data demonstrate that FactorGCL outperforms existing state-of-the-art methods.
  arXiv  Detail & Related papers  (2025-02-05T12:37:15Z)
• STORM: A Spatio-Temporal Factor Model Based on Dual Vector Quantized Variational Autoencoders for Financial Trading [55.02735046724146]
  In financial trading, factor models are widely used to price assets and capture excess returns from mispricing.<n>We propose a Spatio-Temporal factOR Model based on dual vector quantized variational autoencoders, named STORM.<n>Storm extracts features of stocks from temporal and spatial perspectives, then fuses and aligns these features at the fine-grained and semantic level, and represents the factors as multi-dimensional embeddings.
  arXiv  Detail & Related papers  (2024-12-12T17:15:49Z)
• Diffusion Variational Autoencoder for Tackling Stochasticity in Multi-Step Regression Stock Price Prediction [54.21695754082441]
  Multi-step stock price prediction over a long-term horizon is crucial for forecasting its volatility. Current solutions to multi-step stock price prediction are mostly designed for single-step, classification-based predictions. We combine a deep hierarchical variational-autoencoder (VAE) and diffusion probabilistic techniques to do seq2seq stock prediction. Our model is shown to outperform state-of-the-art solutions in terms of its prediction accuracy and variance.
  arXiv  Detail & Related papers  (2023-08-18T16:21:15Z)
• Factor Investing with a Deep Multi-Factor Model [123.52358449455231]
  We develop a novel deep multi-factor model that adopts industry neutralization and market neutralization modules with clear financial insights. Tests on real-world stock market data demonstrate the effectiveness of our deep multi-factor model.
  arXiv  Detail & Related papers  (2022-10-22T14:47:11Z)
• Risk budget portfolios with convex Non-negative Matrix Factorization [0.0]
  We propose a portfolio allocation method based on risk factor budgeting using convex Nonnegative Matrix Factorization (NMF) We evaluate our method in the context of volatility targeting on two long-only global portfolios of cryptocurrencies and traditional assets.
  arXiv  Detail & Related papers  (2022-04-06T12:02:15Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.