RETuning: Upgrading Inference-Time Scaling for Stock Movement Prediction with Large Language Models

• URL: http://arxiv.org/abs/2510.21604v1
• Date: Fri, 24 Oct 2025 16:08:33 GMT
• Title: RETuning: Upgrading Inference-Time Scaling for Stock Movement Prediction with Large Language Models
• Authors: Xueyuan Lin, Cehao Yang, Ye Ma, Ming Li, Rongjunchen Zhang, Yang Ni, Xiaojun Wu, Chengjin Xu, Jian Guo, Hui Xiong,
• Abstract summary: We study a three-class classification problem (up, hold, down) and observe that large language models (LLMs) follow analysts' opinions rather than exhibit a systematic, independent analytical logic (CoTs)<n>We propose Reflective Evidence Tuning (RETuning), a cold-start method prior to reinforcement learning, to enhance prediction ability.<n>We build a large-scale dataset spanning all of 2024 for 5,123 A-share stocks, with long contexts (32K tokens) and over 200K samples.
• Score: 37.97736341087795
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Recently, large language models (LLMs) have demonstrated outstanding reasoning capabilities on mathematical and coding tasks. However, their application to financial tasks-especially the most fundamental task of stock movement prediction-remains underexplored. We study a three-class classification problem (up, hold, down) and, by analyzing existing reasoning responses, observe that: (1) LLMs follow analysts' opinions rather than exhibit a systematic, independent analytical logic (CoTs). (2) LLMs list summaries from different sources without weighing adversarial evidence, yet such counterevidence is crucial for reliable prediction. It shows that the model does not make good use of its reasoning ability to complete the task. To address this, we propose Reflective Evidence Tuning (RETuning), a cold-start method prior to reinforcement learning, to enhance prediction ability. While generating CoT, RETuning encourages dynamically constructing an analytical framework from diverse information sources, organizing and scoring evidence for price up or down based on that framework-rather than on contextual viewpoints-and finally reflecting to derive the prediction. This approach maximally aligns the model with its learned analytical framework, ensuring independent logical reasoning and reducing undue influence from context. We also build a large-scale dataset spanning all of 2024 for 5,123 A-share stocks, with long contexts (32K tokens) and over 200K samples. In addition to price and news, it incorporates analysts' opinions, quantitative reports, fundamental data, macroeconomic indicators, and similar stocks. Experiments show that RETuning successfully unlocks the model's reasoning ability in the financial domain. Inference-time scaling still works even after 6 months or on out-of-distribution stocks, since the models gain valuable insights about stock movement prediction.

Related papers

• Impact of LLMs news Sentiment Analysis on Stock Price Movement Prediction [5.317864735982288]
  This paper addresses stock price movement prediction by leveraging LLM-based news sentiment analysis.<n>We compare 3 different LLMs, namely, DeBERTa, RoBERTa and FinBERT, for sentiment-driven stock prediction.<n>Our results suggest that DeBERTa outperforms the other two models with an accuracy of 75% and that an ensemble model that combines the three models can increase the accuracy to about 80%.
  arXiv  Detail & Related papers  (2026-01-22T23:33:02Z)
• Your AI, Not Your View: The Bias of LLMs in Investment Analysis [62.388554963415906]
  In finance, Large Language Models (LLMs) face frequent knowledge conflicts arising from discrepancies between their pre-trained parametric knowledge and real-time market data.<n>These conflicts are especially problematic in real-world investment services, where a model's inherent biases can misalign with institutional objectives.<n>We propose an experimental framework to investigate emergent behaviors in such conflict scenarios, offering a quantitative analysis of bias in investment analysis.
  arXiv  Detail & Related papers  (2025-07-28T16:09:38Z)
• Context is Key: A Benchmark for Forecasting with Essential Textual Information [87.3175915185287]
  "Context is Key" (CiK) is a forecasting benchmark that pairs numerical data with diverse types of carefully crafted textual context.<n>We evaluate a range of approaches, including statistical models, time series foundation models, and LLM-based forecasters.<n>We propose a simple yet effective LLM prompting method that outperforms all other tested methods on our benchmark.
  arXiv  Detail & Related papers  (2024-10-24T17:56:08Z)
• Harnessing Earnings Reports for Stock Predictions: A QLoRA-Enhanced LLM Approach [6.112119533910774]
  This paper introduces an advanced approach by employing Large Language Models (LLMs) instruction fine-tuned with a novel combination of instruction-based techniques and quantized low-rank adaptation (QLoRA) compression. Our methodology integrates 'base factors', such as financial metric growth and earnings transcripts, with 'external factors', including recent market indices performances and analyst grades, to create a rich, supervised dataset. This study not only demonstrates the power of integrating cutting-edge AI with fine-tuned financial data but also paves the way for future research in enhancing AI-driven financial analysis tools.
  arXiv  Detail & Related papers  (2024-08-13T04:53:31Z)
• ECC Analyzer: Extract Trading Signal from Earnings Conference Calls using Large Language Model for Stock Performance Prediction [7.358590821647365]
  This research introduces a novel framework: textbfECC Analyzer, which utilizes large language models (LLMs) to extract richer, more predictive content from ECCs. We use the pre-trained large models to extract textual and audio features from ECCs and implement a hierarchical information extraction strategy to extract more fine-grained information. Experimental results demonstrate that our model outperforms traditional analytical benchmarks.
  arXiv  Detail & Related papers  (2024-04-29T07:11:39Z)
• AlphaFin: Benchmarking Financial Analysis with Retrieval-Augmented Stock-Chain Framework [48.3060010653088]
  We release AlphaFin datasets, combining traditional research datasets, real-time financial data, and handwritten chain-of-thought (CoT) data. We then use AlphaFin datasets to benchmark a state-of-the-art method, called Stock-Chain, for effectively tackling the financial analysis task.
  arXiv  Detail & Related papers  (2024-03-19T09:45:33Z)
• Ploutos: Towards interpretable stock movement prediction with financial large language model [43.51934592920784]
  Ploutos is a novel financial framework that consists of PloutosGen and PloutosGPT. The PloutosGen contains multiple primary experts that can analyze different modal data, such as text and numbers, and provide quantitative strategies from different perspectives. The training strategy of PloutosGPT leverage rearview-mirror prompting mechanism to guide GPT-4 to generate rationales, and a dynamic token weighting mechanism to finetune LLM.
  arXiv  Detail & Related papers  (2024-02-18T10:28:18Z)
• Learning to Generate Explainable Stock Predictions using Self-Reflective Large Language Models [54.21695754082441]
  We propose a framework to teach Large Language Models (LLMs) to generate explainable stock predictions. A reflective agent learns how to explain past stock movements through self-reasoning, while the PPO trainer trains the model to generate the most likely explanations. Our framework can outperform both traditional deep-learning and LLM methods in prediction accuracy and Matthews correlation coefficient.
  arXiv  Detail & Related papers  (2024-02-06T03:18:58Z)
• Can ChatGPT Forecast Stock Price Movements? Return Predictability and Large Language Models [51.3422222472898]
  We document the capability of large language models (LLMs) like ChatGPT to predict stock price movements using news headlines. We develop a theoretical model incorporating information capacity constraints, underreaction, limits-to-arbitrage, and LLMs.
  arXiv  Detail & Related papers  (2023-04-15T19:22:37Z)
• Stock Broad-Index Trend Patterns Learning via Domain Knowledge Informed Generative Network [2.1163070161951865]
  We propose IndexGAN, which includes deliberate designs for the inherent characteristics of the stock market. We also utilize the critic to approximate the Wasserstein distance between actual and predicted sequences.
  arXiv  Detail & Related papers  (2023-02-27T21:56:56Z)
• Value-driven Hindsight Modelling [68.658900923595]
  Value estimation is a critical component of the reinforcement learning (RL) paradigm. Model learning can make use of the rich transition structure present in sequences of observations, but this approach is usually not sensitive to the reward function. We develop an approach for representation learning in RL that sits in between these two extremes. This provides tractable prediction targets that are directly relevant for a task, and can thus accelerate learning the value function.
  arXiv  Detail & Related papers  (2020-02-19T18:10:20Z)

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.