A Survey of Large Language Models for Financial Applications: Progress, Prospects and Challenges

Paper: arXiv 2406.11903

Abstract

Recent advances in large language models (LLMs) have unlocked novel opportunities for machine learning applications in the financial domain. These models have demonstrated remarkable capabilities in understanding context, processing vast amounts of data, and generating human-preferred contents. In this survey, we explore the application of LLMs on various financial tasks, focusing on their potential to transform traditional practices and drive innovation. We provide a discussion of the progress and advantages of LLMs in financial contexts, analyzing their advanced technologies as well as prospective capabilities in contextual understanding, transfer learning flexibility, complex emotion detection, etc. We then highlight this survey for categorizing the existing literature into key application areas, including linguistic tasks, sentiment analysis, financial time series, financial reasoning, agent-based modeling, and other applications. For each application area, we delve into specific methodologies, such as textual analysis, knowledge-based analysis, forecasting, data augmentation, planning, decision support, and simulations. Furthermore, a comprehensive collection of datasets, model assets, and useful codes associated with mainstream applications are presented as resources for the researchers and practitioners. Finally, we outline the challenges and opportunities for future research, particularly emphasizing a number of distinctive aspects in this field. We hope our work can help facilitate the adoption and further development of LLMs in the financial sector.

Complexity vs Empirical Score

• Math Complexity: 3.5/10
• Empirical Rigor: 6.0/10
• Quadrant: Street Traders — practical and empirical, lighter on theory

Why this score: The paper is a survey review that covers a wide range of topics and datasets but presents minimal novel mathematical derivations or formulas, focusing instead on application and implementation resources. It discusses backtesting challenges and data issues, indicating practical implementation considerations.

Research Flowchart

  flowchart TD
  subgraph A ["Research Goal"]
      direction LR
      A1["Research Goal: Application of LLMs in Finance"]
  end

  subgraph B ["Key Methodology Areas"]
      direction LR
      B1["Linguistic Tasks"]
      B2["Sentiment Analysis"]
      B3["Financial Time Series"]
      B4["Financial Reasoning"]
      B5["Agent-based Modeling"]
  end

  subgraph C ["Data & Inputs"]
      direction LR
      C1["Financial Texts<br/>(News, Reports)"]
      C2["Structured Data<br/>(Market Prices, Stats)"]
      C3["Domain Knowledge<br/>(Finance Rules/Events)"]
  end

  subgraph D ["Computational Processes"]
      direction LR
      D1["Knowledge-based Analysis<br/>& Textual Processing"]
      D2["Data Augmentation<br/>& Reasoning"]
      D3["Forecasting<br/>& Decision Support"]
      D4["Simulations<br/>& Planning"]
  end

  subgraph E ["Key Findings & Outcomes"]
      direction LR
      E1["Prospects: Contextual Understanding<br/>Transfer Learning Flexibility"]
      E2["Challenges: Hallucination<br/>Data Privacy<br/>Temporal Dynamics"]
      E3["Resources: Datasets<br/>Models<br/>Codes"]
  end

  A --> B
  B --> C
  C --> D
  D --> E

Cited References (324)

1. BloombergGPT : A large language model for finance (2023) — link
2. A medical multimodal large language model for future pandemics (2023)
3. WeaverBird : Empowering financial decision-making with large language model, knowledge base, and search engine (2023) — link
4. Large Language Models for Education: A Survey and Outlook (2024) — link
5. Neural network is heterogeneous: Phase matters more (2021) — link
6. Energy Efficiency of Microring Resonator (MRR) -Based Binary Decision Diagram (BDD) Circuits (2019)
7. Improving language understanding by generative pre-training (2018)
8. Language models are unsupervised multitask learners (2019)

32 of 324 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.