# PortBench: A Correlation-Aware, Full-Pipeline Benchmark for LLM-Driven Portfolio Management

arXiv: https://arxiv.org/abs/2605.27887 · HTML: https://mlsanity.com/paper/2605.27887/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 47 references that are in the map.
- First posted on arXiv 27 May 2026.

## Builds on (most cited first, 21 of 21)

- [DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence](https://mlsanity.com/paper/2606.19348/) (2026)
- [GLM-5: from Vibe Coding to Agentic Engineering](https://mlsanity.com/paper/2602.15763/) (2026)
- [FinQA: A Dataset of Numerical Reasoning over Financial Data](https://mlsanity.com/paper/2109.00122/) (2021)
- [FinanceBench: A New Benchmark for Financial Question Answering](https://mlsanity.com/paper/2311.11944/) (2023)
- [PIXIU: A Large Language Model, Instruction Data and Evaluation Benchmark for Finance](https://mlsanity.com/paper/2306.05443/) (2023)
- [ConvFinQA: Exploring the Chain of Numerical Reasoning in Conversational Finance Question Answering](https://mlsanity.com/paper/2210.03849/) (2022)
- [TradingAgents: Multi-Agents LLM Financial Trading Framework](https://mlsanity.com/paper/2412.20138/) (2024)
- [A Multimodal Foundation Agent for Financial Trading: Tool-Augmented, Diversified, and Generalist](https://mlsanity.com/paper/2402.18485/) (2024)
- [WHEN FLUE MEETS FLANG: Benchmarks and Large Pre-trained Language Model for Financial Domain](https://mlsanity.com/paper/2211.00083/) (2022)
- [INVESTORBENCH: A Benchmark for Financial Decision-Making Tasks with LLM-based Agent](https://mlsanity.com/paper/2412.18174/) (2024)
- [R&D-Agent-Quant: A Multi-Agent Framework for Data-Centric Factors and Model Joint Optimization](https://mlsanity.com/paper/2505.15155/) (2025)
- [FinMME: Benchmark Dataset for Financial Multi-Modal Reasoning Evaluation](https://mlsanity.com/paper/2505.24714/) (2025)
- [MarS: a Financial Market Simulation Engine Powered by Generative Foundation Model](https://mlsanity.com/paper/2409.07486/) (2024)
- [FinanceReasoning: Benchmarking Financial Numerical Reasoning More Credible, Comprehensive and Challenging](https://mlsanity.com/paper/2506.05828/) (2025)
- [HedgeAgents: A Balanced-aware Multi-agent Financial Trading System](https://mlsanity.com/paper/2502.13165/) (2025)
- [Time Travel is Cheating: Going Live with DeepFund for Real-Time Fund Investment Benchmarking](https://mlsanity.com/paper/2505.11065/) (2025)
- [LLM-Powered Multi-Agent System for Automated Crypto Portfolio Management](https://mlsanity.com/paper/2501.00826/) (2025)
- [Standard Benchmarks Fail -- Auditing LLM Agents in Finance Must Prioritize Risk](https://mlsanity.com/paper/2502.15865/) (2025)
- [FinDABench: Benchmarking Financial Data Analysis Ability of Large Language Models](https://mlsanity.com/paper/2401.02982/) (2024)
- [MASS: Muli-agent simulation scaling for portfolio construction](https://mlsanity.com/paper/2505.10278/) (2025)
- [FinRipple: Aligning Large Language Models with Financial Market for Event Ripple Effect Awareness](https://mlsanity.com/paper/2505.23826/) (2025)

## Built on by (most cited first, 3 of 3)

- [CLQT: A Closed-Loop, Cost-Aware, Strategy-Consistent Benchmark for Diagnostic Evaluation of LLM Portfolio-Management Agents](https://mlsanity.com/paper/2606.29771/) (2026)
- [Beyond Agent Architecture: Execution Assumptions and Reproducibility in LLM-Based Trading Systems](https://mlsanity.com/paper/2606.08285/) (2026)
- [MemTrial: Learning When to Trust Memory in LLM Portfolio Agents](https://mlsanity.com/paper/2610.11732/) (2026)

## Abstract (arXiv)

Large language models (LLMs) have shown strong performance across diverse financial tasks, yet portfolio management (PM) remains poorly evaluated. Existing benchmarks exhibit two gaps: they are often equity-only and ignore cross-asset correlations; they provide limited insight into where errors arise and how they affect subsequent decisions. We introduce PortBench, a benchmark with market data spanning six asset classes from 2015 to 2025. PortBench comprises a static QA dataset of 6,269 questions across seven task templates and a dynamic five-stage portfolio evaluation. Dynamic evaluation uses a class-level correlation score for allocation and CEPS to summarize reference-based stage scores. We further evaluate under three stress windows and three risk profiles, and support real-time evaluation on newly arriving market data. Evaluating ten LLMs reveals a disconnect: high financial QA scores do not ensure consistent portfolio outperformance. LLMs concentrate exposure in correlated assets, and their portfolios achieve higher Sharpe than equal weighting in 54.2% of 120 evaluations across four market periods. Our source code is available at hrefhttps://github.com/AgenticFinLab/portbenchthis https URL.