Paper: arXiv 2508.11152
Authors: Tianjiao Zhao, Jingrao Lyu, Stokes Jones, Harrison Garber, Stefano Pasquali, Dhagash Mehta
Abstract
The field of artificial intelligence (AI) agents is evolving rapidly, driven by the capabilities of Large Language Models (LLMs) to autonomously perform and refine tasks with human-like efficiency and adaptability. In this context, multi-agent collaboration has emerged as a promising approach, enabling multiple AI agents to work together to solve complex challenges. This study investigates the application of role-based multi-agent systems to support stock selection in equity research and portfolio management. We present a comprehensive analysis performed by a team of specialized agents and evaluate their stock-picking performance against established benchmarks under varying levels of risk tolerance. Furthermore, we examine the advantages and limitations of employing multi-agent frameworks in equity analysis, offering critical insights into their practical efficacy and implementation challenges.
Complexity vs Empirical Score
- Math Complexity: 2.5/10
- Empirical Rigor: 6.0/10
- Quadrant: Street Traders — empirical and applied, lower math complexity
Why this score: The paper relies primarily on LLM prompt engineering and system design rather than advanced mathematical derivations, resulting in low math complexity; however, it includes a systematic backtest against benchmarks with specific performance metrics and risk profiles, demonstrating substantial empirical rigor.
Research Flowchart
flowchart TD A["Research Goal: LLM Multi-Agents for Equity Portfolio Construction"] --> B["Methodology: Role-Based Multi-Agent System"] B --> C["Inputs: Financial Data & Market News"] C --> D["Process: Collaborative Stock Selection"] D --> E["Process: Portfolio Construction vs Benchmarks"] E --> F["Outcomes: Stock-Picking Performance & Risk Analysis"] F --> G["Conclusion: Efficacy & Challenges of Multi-Agent Frameworks"]
Evidence Card
What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card
Proposes
The paper proposes a novel multi‑agent investment framework that uses LLM‑powered collaboration and debate to collect and synthesize equity research inputs.
Source sentence
In this work, we propose a novel multi-agent investment framework that leverages LLM-powered collaboration and debate to collect and synthesize equity research inputs.
Assumes 2
- Each agent only accesses data relevant to its specific role and assigned task. data
Source sentence
Each agent is granted access only to the data relevant to its specific role and assigned task.
- Risk‑averse and risk‑neutral agent behavior is modeled via prompt engineering that embeds investor traits into agent instructions. model
Source sentence
We design agents for both risk‑averse and risk‑neutral scenarios using prompt engineering to embed investor traits directly into agent instructions.
Data
2024-02 – 2024-06 · equities
- 10‑K and 10‑Q financial disclosures fundamentals public
Source sentence
10K/10Q Fundamental Agent Financial disclosure data (includes discussion of major company prospectus and financial statements).
- Bloomberg financial news text-news commercial
Source sentence
Bloomberg ID Sentiment Agent Ticker’s corresponding unique identifier for Bloomberg News
- Historical stock price and volume data market-prices public
Source sentence
Price (Open, High, Low, Close) Valuation Agent The initial, highest, lowest, and final trading price of a stock within a specific trading period
Costs
Not modeled
Finds 3
- In the risk‑neutral setting, the multi‑agent framework outperforms both the benchmark and single‑agent frameworks during the testing window. performance
Source sentence
In the risk-neutral setting, the proposed multi-agent outperforms the benchmark as well as the single agent framework during the testing window.
- In the risk‑averse setting, all agent‑selected portfolios underperform the benchmark, partly due to exclusion of volatile technology stocks. performance
Source sentence
In the risk-averse setting, all agent-selected portfolios followed a conservative investment strategy, leading to underperformance relative to the benchmark.
- The multi‑agent portfolio shows slightly lower volatility and reduced drawdowns compared to individual valuation and fundamental portfolios. performance
Source sentence
the multi-agent portfolio exhibited slightly lower volatility and reduced drawdowns, indicating an ability to mitigate losses more effectively than the individual strategies.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 22 related papers). Coverage is partial, so this is neither replication nor refutation.
8 papers in the archive
Code
No repository by the authors found in the paper.
Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.
Cited References (125)
- MAPS: Multi-agent reinforcement learning-based portfolio management system (2020) — link
- Designing a multi-agent portfolio management system (1996)
- Evaluating multi-agent coordination abilities in large language models (2023) — link
- Multi-agent deep reinforcement learning algorithm with trend consistency regularization for portfolio management (2023)
- FinMem: A Performance-Enhanced LLM Trading Agent with Layered Memory and Character Design — → 2311.13743
- A Multimodal Foundation Agent for Financial Trading: Tool-Augmented, Diversified, and Generalist — → 2402.18485
- Self-Reflection in LLM Agents: Effects on Problem-Solving Performance (2024) — link
- Emotional finance: The role of the unconscious in financial decisions (2010)
and 117 more
- Behavioral finance and portfolio management: Review of theory and literature (2020)
- Applications of Client Behavior: A Practitioner’s Perspective (2017)
- Behavioral finance-theory and practical application: Systematic analysis of departures from the homo oeconomicus paradigm are essential for realistic financial research and analysis (2001)
- Multi-agent collaboration: Harnessing the power of intelligent llm agents (2023) — link
- Herding in the Italian stock market: a case of behavioral finance (2004)
- Financial decision-making in markets and firms: A behavioral perspective (1995)
- Practical Challenges of Implementing Behavioral Finance: Reflections from the Field (2017)
- Behavioral finance: investors, corporations, and markets (2010)
- Prospect theory: An analysis of decision under risk (2013)
- Autogen: Enabling next-gen llm applications via multi-agent conversation framework (2023) — link
- More agents is all you need (2024) — link
- Marketsenseai 2.0: Enhancing stock analysis through llm agents — → 2502.00415
- Llm-planner: Few-shot grounded planning for embodied agents with large language models (2023)
- Api-bank: A comprehensive benchmark for tool-augmented llms (2023) — link
- Towards Efficient LLM Grounding for Embodied Multi-Agent Collaboration (2024) — link
- Medagents: Large language models as collaborators for zero-shot medical reasoning (2023) — link
- Communicative agents for software development (2023) — link
- Metagpt: Meta programming for multi-agent collaborative framework (2023) — link
- A survey on in-context learning (2022) — link
- Impact of risk tolerance and demographic factors on financial investment decision (2018)
- FinRobot: An Open-Source AI Agent Platform for Financial Applications using Large Language Models — → 2405.14767
- Llm-based nlg evaluation: Current status and challenges (2024) — link
- Retrieval-augmented generation for large language models: A survey (2023) — link
- Videoagent: Long-form video understanding with large language model as agent (2024) — link
- Bloomberggpt: A large language model for finance (2023) — link
- Stock ranking prediction using list-wise approach and node embedding technique (2021)
- Minilm: Deep self-attention distillation for task-agnostic compression of pre-trained transformers (2020)
- Universal language model fine-tuning for text classification (2018) — link
- Improving factuality and reasoning in language models through multiagent debate (2023) — link
- Attention is all you need (2017)
- Bert: Pre-training of deep bidirectional transformers for language understanding (2018) — link
- Recommender systems (2016)
- Categorizing mutual funds using clusters (1999)
- Scikit-learn: Machine learning in Python (2011)
- “Morningstar Categorization.” (2018) — link
- Recommender systems (1997)
- Clustering financial time series: an application to mutual funds style analysis (2004)
- Mutual funds apart from the crowd (2018)
- Mutual fund objective misclassification (2000)
- Time series clustering and classification by the autoregressive metric (2008)
- Soft Clustering for Funds Management Style Analysis: Out-of-Sample Predictability (2008)
- Fund2Vec: Mutual Funds Similarity using Graph Learning (2021) — link
- Machine learning fund categorizations (2020)
- On Robustness of Mutual Funds Categorization and Distance Metric Learning (2021)
- node2vec: Scalable feature learning for networks (2016)
- Does Mutual Fund Management in India correspond to its investment objective classification? (2005)
- Compensation incentives and risk taking behavior: evidence from mutual funds (1996)
- Mutual fund misclassification: Evidence based on style analysis (1997)
- Incentive fees and mutual funds (2003)
- The Consistency of Self-Declared Hedge Fund Styles—A Return-Based Analysis with Self-Organizing Maps (2005)
- Mutual fund styles (1997)
- Tweets and peers: defining industry groups and strategic peers based on investor perceptions of stocks on Twitter (2011)
- Patent citations and stock performance: Constructing a dynamic industry classification (2014)
- Can we design an industry classification system that reflects industry architecture? (2018)
- Search-based peer firms: Aggregating investor perceptions through internet co-searches (2015)
- Text-based network industries and endogenous product differentiation (2016)
- Introducing the North American industry classification system (1998)
- Global industry classification standard (GICS \textregistered ) methodology: guiding principles and methodology for GICS (2020)
- Industry costs of equity (1997)
- Social learning and corporate peer effects (2015)
- Differences between COMPUSTAT and CRSP SIC codes and related effects on research (1994)
- Spanish Mutual Fund Misclassification: Empirical Evidence (2005)
- Stylistic Differences across Hedge Funds as Revealed by Historical Monthly Returns (2010)
- hedge Fund classification using K-means clustering Method (2003)
- Ultrametricity in fund of funds diversification (2004)
- Cluster analysis and manager selection (1986)
- Stable Return-Based Fund Classification (2017)
- Rifinitive. “Lipper Fund Research.” (2016) — link
- Classifying companies by industry using word embeddings (2018)
- Clustering mutual funds by return and risk levels (2010)
- Variance-based criteria for clustering and their application to the analysis of management styles of mutual funds based on time series of daily returns (2008)
- Using Risk Characteristics to Classify Funds (2018)
- Objectives and performance of mutual funds, 1960–1969 (1974)
- Asset allocation: Management style and performance measurement (1992)
- Visualizing stock-mutual fund relationships through social network analysis (2009)
- Clustering approaches for financial data analysis: a survey (2016) — link
- Do fund objectives affect diversification policies? (1982)
- Classifying mutual funds in India: Some results from clustering (2007)
- Related stocks selection with data collaboration using text mining (2019)
- The style consistency of hedge funds (2007)
- A Data-Driven Categorization of Investable Assets (2015)
- Clustering mutual funds based on investment similarity (2015)
- Self-organizing maps could improve the classification of Spanish mutual funds (2006)
- Morningstar’s classification of large-cap mutual funds (2001)
- Double clustering for rating mutual funds (2015)
- Distributed Representations of Sentences and Documents (2014) — link
- Learning Mutual Fund Categorization using Natural Language Processing (2022)
- Don’t Take Their Word for It: The Misclassification of Bond Mutual Funds (2021)
- Changing names with style: Mutual fund name changes and their effects on fund flows (2005)
- Investment style misclassification and mutual fund performance (2017)
- Efficient Estimation of Word Representations in Vector Space (2013) — link
- Longformer: The long-document transformer (2020) — link
- Palm: Scaling language modeling with pathways (2022) — link
- Overcoming catastrophic forgetting in neural networks (2017)
- Language models are few-shot learners (2020)
- Palm: Scaling language modeling with pathways (2022) — link
- Mpnet: Masked and permuted pre-training for language understanding (2020)
- Language models are unsupervised multitask learners (2019)
- Pathways: Asynchronous distributed dataflow for ml (2022)
- Roberta: A robustly optimized bert pretraining approach (2019) — link
- Instance-based Explanations for Gradient Boosting Machine Predictions with AXIL Weights (2023) — link
- How Trading Analytics and Data Science Can Improve Investment Outcomes (2022)
- Liquidity—how to capture a multidimensional beast (2016)
- Geometry- and Accuracy-Preserving Random Forest Proximities (2023) — link
- Quantifying Outlierness of Funds from their Categories using Supervised Similarity — → 2308.06882
- Supervised similarity learning for corporate bonds using Random Forest proximities (2022)
- Random Forests (2001)
- A unified approach to interpreting model predictions (2017)
- Discriminatory analysis. Nonparametric discrimination: Consistency properties (1989)
- The Handbook of Fixed Income Securities, Eighth Edition (2011) — link
- A Survey of Large Language Models for Financial Applications: Progress, Prospects and Challenges — → 2406.11903
- Reinforcement learning: An introduction (2018)
- A comprehensive overview of large language models (2023) — link
- A survey on large language model based autonomous agents (2024)
- Large language model based multi-agents: A survey of progress and challenges (2024) — link
- A systematic survey of prompt engineering in large language models: Techniques and applications (2024) — link
- FinVerse: An Autonomous Agent System for Versatile Financial Analysis (2024) — link
6 of 125 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.