InvestLM: A Large Language Model for Investment using Financial Domain Instruction Tuning

Paper: arXiv 2309.13064

Authors: Yi Yang, Yixuan Tang, Kar Yan Tam

Abstract

We present a new financial domain large language model, InvestLM, tuned on LLaMA-65B (Touvron et al., 2023), using a carefully curated instruction dataset related to financial investment. Inspired by less-is-more-for-alignment (Zhou et al., 2023), we manually curate a small yet diverse instruction dataset, covering a wide range of financial related topics, from Chartered Financial Analyst (CFA) exam questions to SEC filings to Stackexchange quantitative finance discussions. InvestLM shows strong capabilities in understanding financial text and provides helpful responses to investment related questions. Financial experts, including hedge fund managers and research analysts, rate InvestLM’s response as comparable to those of state-of-the-art commercial models (GPT-3.5, GPT-4 and Claude-2). Zero-shot evaluation on a set of financial NLP benchmarks demonstrates strong generalizability. From a research perspective, this work suggests that a high-quality domain specific LLM can be tuned using a small set of carefully curated instructions on a well-trained foundation model, which is consistent with the Superficial Alignment Hypothesis (Zhou et al., 2023). From a practical perspective, this work develops a state-of-the-art financial domain LLM with superior capability in understanding financial texts and providing helpful investment advice, potentially enhancing the work efficiency of financial professionals. We release the model parameters to the research community.

Complexity vs Empirical Score

• Math Complexity: 2.0/10
• Empirical Rigor: 6.5/10
• Quadrant: Street Traders — practical and empirical, lighter on theory

Why this score: InvestLM provides a practical financial LLM application by applying instruction tuning on a LLaMA base model using a curated dataset. While mathematical complexity is minimal, the paper features solid empirical evaluation using expert human ratings (hedge fund managers/analysts) and standard NLP benchmarks, alongside open-weight release for reproducibility.

Research Flowchart

  flowchart TD
    A[Research Goal: Develop SOTA Financial LLM via Instruction Tuning] --> B[Data Curation: CFA questions, SEC filings & QFin StackExchange]
    B --> C[Instruction Tuning: Fine-tune LLaMA-65B Foundation Model]
    C --> D[Evaluation Phase]
    D --> E[Human Expert Rating: Hedge fund managers & analysts compare to GPT-4/Claude-2]
    D --> F[Zero-Shot NLP Benchmarks: Test generalizability]
    E --> G[Outcomes: Comparable to SOTA commercial LLMs & confirms Superficial Alignment Hypothesis]
    F --> G