Agent Trading Arena: A Study on Numerical Understanding in LLM-Based Agents

• URL: http://arxiv.org/abs/2502.17967v2
• Date: Tue, 02 Sep 2025 02:28:24 GMT
• Title: Agent Trading Arena: A Study on Numerical Understanding in LLM-Based Agents
• Authors: Tianmi Ma, Jiawei Du, Wenxin Huang, Wenjie Wang, Liang Xie, Xian Zhong, Joey Tianyi Zhou,
• Abstract summary: Large language models (LLMs) have demonstrated remarkable capabilities in natural language tasks.<n>We present the Agent Trading Arena, a virtual zero-sum stock market in which LLM-based agents engage in competitive multi-agent trading.
• Score: 69.58565132975504
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models (LLMs) have demonstrated remarkable capabilities in natural language tasks, yet their performance in dynamic, real-world financial environments remains underexplored. Existing approaches are limited to historical backtesting, where trading actions cannot influence market prices and agents train only on static data. To address this limitation, we present the Agent Trading Arena, a virtual zero-sum stock market in which LLM-based agents engage in competitive multi-agent trading and directly impact price dynamics. By simulating realistic bid-ask interactions, our platform enables training in scenarios that closely mirror live markets, thereby narrowing the gap between training and evaluation. Experiments reveal that LLMs struggle with numerical reasoning when given plain-text data, often overfitting to local patterns and recent values. In contrast, chart-based visualizations significantly enhance both numerical reasoning and trading performance. Furthermore, incorporating a reflection module yields additional improvements, especially with visual inputs. Evaluations on NASDAQ and CSI datasets demonstrate the superiority of our method, particularly under high volatility. All code and data are available at https://github.com/wekjsdvnm/Agent-Trading-Arena.

Related papers

• LiveTradeBench: Seeking Real-World Alpha with Large Language Models [26.976122048323873]
  Large language models (LLMs) achieve strong performance across benchmarks.<n>These tests occur in static settings, lacking real dynamics and uncertainty.<n>We introduce LiveTradeBench, a live trading environment for evaluating LLM agents in realistic and evolving markets.
  arXiv  Detail & Related papers  (2025-11-05T16:47:26Z)
• See, Think, Act: Online Shopper Behavior Simulation with VLM Agents [58.92444959954643]
  This paper investigates the integration of visual information, specifically webpage screenshots, into behavior simulation via VLMs.<n>We employ SFT for joint action prediction and rationale generation, conditioning on the full interaction context.<n>To further enhance reasoning capabilities, we integrate RL with a hierarchical reward structure, scaled by a difficulty-aware factor.
  arXiv  Detail & Related papers  (2025-10-22T05:07:14Z)
• When Agents Trade: Live Multi-Market Trading Benchmark for LLM Agents [74.55061622246824]
  Agent Market Arena (AMA) is the first lifelong, real-time benchmark for evaluating Large Language Model (LLM)-based trading agents.<n>AMA integrates verified trading data, expert-checked news, and diverse agent architectures within a unified trading framework.<n>It evaluates agents across GPT-4o, GPT-4.1, Claude-3.5-haiku, Claude-sonnet-4, and Gemini-2.0-flash.
  arXiv  Detail & Related papers  (2025-10-13T17:54:09Z)
• On the Performance of LLMs for Real Estate Appraisal [5.812129569528997]
  This study examines how Large Language Models (LLMs) can democratize access to real estate insights by generating competitive and interpretable house price estimates.<n>We evaluate leading LLMs on diverse international housing datasets, comparing zero-shot, few-shot, market report-enhanced, and hybrid prompting techniques.<n>Our results show that LLMs effectively leverage hedonic variables, such as property size and amenities, to produce meaningful estimates.
  arXiv  Detail & Related papers  (2025-06-13T14:14:40Z)
• Iterative Self-Incentivization Empowers Large Language Models as Agentic Searchers [74.17516978246152]
  Large language models (LLMs) have been widely integrated into information retrieval to advance traditional techniques.<n>We propose EXSEARCH, an agentic search framework, where the LLM learns to retrieve useful information as the reasoning unfolds.<n>Experiments on four knowledge-intensive benchmarks show that EXSEARCH substantially outperforms baselines.
  arXiv  Detail & Related papers  (2025-05-26T15:27:55Z)
• IDA-Bench: Evaluating LLMs on Interactive Guided Data Analysis [60.32962597618861]
  IDA-Bench is a novel benchmark evaluating large language models in multi-round interactive scenarios.<n>Agent performance is judged by comparing its final numerical output to the human-derived baseline.<n>Even state-of-the-art coding agents (like Claude-3.7-thinking) succeed on 50% of the tasks, highlighting limitations not evident in single-turn tests.
  arXiv  Detail & Related papers  (2025-05-23T09:37:52Z)
• Scaling Autonomous Agents via Automatic Reward Modeling And Planning [52.39395405893965]
  Large language models (LLMs) have demonstrated remarkable capabilities across a range of tasks.<n>However, they still struggle with problems requiring multi-step decision-making and environmental feedback.<n>We propose a framework that can automatically learn a reward model from the environment without human annotations.
  arXiv  Detail & Related papers  (2025-02-17T18:49:25Z)
• Mitigating Forgetting in LLM Fine-Tuning via Low-Perplexity Token Learning [61.99353167168545]
  We show that fine-tuning with LLM-generated data improves target task performance and reduces non-target task degradation.<n>This is the first work to provide an empirical explanation based on token perplexity reduction to mitigate catastrophic forgetting in LLMs after fine-tuning.
  arXiv  Detail & Related papers  (2025-01-24T08:18:56Z)
• Leveraging Online Olympiad-Level Math Problems for LLMs Training and Contamination-Resistant Evaluation [55.21013307734612]
  AoPS-Instruct is a dataset of more than 600,000 high-quality QA pairs.<n>LiveAoPSBench is an evolving evaluation set with timestamps, derived from the latest forum data.<n>Our work presents a scalable approach to creating and maintaining large-scale, high-quality datasets for advanced math reasoning.
  arXiv  Detail & Related papers  (2025-01-24T06:39:38Z)
• TradingAgents: Multi-Agents LLM Financial Trading Framework [4.293484524693143]
  TradingAgents proposes a novel stock trading framework inspired by trading firms.<n>It features LLM-powered agents in specialized roles such as fundamental analysts, sentiment analysts, technical analysts, and traders with varied risk profiles.<n>By simulating a dynamic, collaborative trading environment, this framework aims to improve trading performance.
  arXiv  Detail & Related papers  (2024-12-28T12:54:06Z)
• Invar-RAG: Invariant LLM-aligned Retrieval for Better Generation [43.630437906898635]
  We propose a novel two-stage fine-tuning architecture called Invar-RAG. In the retrieval stage, an LLM-based retriever is constructed by integrating LoRA-based representation learning. In the generation stage, a refined fine-tuning method is employed to improve LLM accuracy in generating answers based on retrieved information.
  arXiv  Detail & Related papers  (2024-11-11T14:25:37Z)
• Distill Visual Chart Reasoning Ability from LLMs to MLLMs [38.62832112530892]
  Solving complex chart Q&A tasks requires advanced visual reasoning abilities in multimodal large language models (MLLMs) We propose Code-as-Intermediary Translation (CIT), a cost-effective, efficient and easily scalable data synthesis method for distilling visual reasoning abilities from LLMs to MLLMs. We employ text-based synthesizing techniques to construct chart-plotting code and produce ReachQA, a dataset containing 3k reasoning-intensive charts and 20k Q&A pairs.
  arXiv  Detail & Related papers  (2024-10-24T14:50:42Z)
• When AI Meets Finance (StockAgent): Large Language Model-based Stock Trading in Simulated Real-world Environments [55.19252983108372]
  We have developed a multi-agent AI system called StockAgent, driven by LLMs. The StockAgent allows users to evaluate the impact of different external factors on investor trading. It avoids the test set leakage issue present in existing trading simulation systems based on AI Agents.
  arXiv  Detail & Related papers  (2024-07-15T06:49:30Z)
• DARG: Dynamic Evaluation of Large Language Models via Adaptive Reasoning Graph [70.79413606968814]
  We introduce Dynamic Evaluation of LLMs via Adaptive Reasoning Graph Evolvement (DARG) to dynamically extend current benchmarks with controlled complexity and diversity. Specifically, we first extract the reasoning graphs of data points in current benchmarks and then perturb the reasoning graphs to generate novel testing data. Such newly generated test samples can have different levels of complexity while maintaining linguistic diversity similar to the original benchmarks.
  arXiv  Detail & Related papers  (2024-06-25T04:27:53Z)
• ChatGLM-Math: Improving Math Problem-Solving in Large Language Models with a Self-Critique Pipeline [42.61538071832468]
  Large language models (LLMs) have shown excellent mastering of human language, but still struggle in real-world applications that require mathematical problem-solving. We tailor the Self-Critique pipeline, which addresses the challenge in the feedback learning stage of LLM alignment.
  arXiv  Detail & Related papers  (2024-04-03T17:51:18Z)
• Unsupervised Information Refinement Training of Large Language Models for Retrieval-Augmented Generation [128.01050030936028]
  We propose an information refinement training method named InFO-RAG. InFO-RAG is low-cost and general across various tasks. It improves the performance of LLaMA2 by an average of 9.39% relative points.
  arXiv  Detail & Related papers  (2024-02-28T08:24:38Z)
• An Auction-based Marketplace for Model Trading in Federated Learning [54.79736037670377]
  Federated learning (FL) is increasingly recognized for its efficacy in training models using locally distributed data. We frame FL as a marketplace of models, where clients act as both buyers and sellers. We propose an auction-based solution to ensure proper pricing based on performance gain.
  arXiv  Detail & Related papers  (2024-02-02T07:25:53Z)
• Data-Centric Financial Large Language Models [27.464319154543173]
  Large language models (LLMs) show promise for natural language tasks but struggle when applied directly to complex domains like finance. We propose a data-centric approach to enable LLMs to better handle financial tasks.
  arXiv  Detail & Related papers  (2023-10-07T04:53:31Z)
• Neural Stochastic Agent-Based Limit Order Book Simulation: A Hybrid Methodology [6.09170287691728]
  Modern financial exchanges use an electronic limit order book (LOB) to store bid and ask orders for a specific financial asset. We propose a novel hybrid LOB simulation paradigm characterised by: (1) representing the aggregation of market events' logic by a neural background trader that is pre-trained on historical LOB data through a neural point model; and (2) embedding the background trader in a multi-agent simulation with other trading agents. We show that the stylised facts remain and we demonstrate order flow impact and financial herding behaviours that are in accordance with empirical observations of real markets.
  arXiv  Detail & Related papers  (2023-02-28T20:53:39Z)
• Check Your Facts and Try Again: Improving Large Language Models with External Knowledge and Automated Feedback [127.75419038610455]
  Large language models (LLMs) are able to generate human-like, fluent responses for many downstream tasks. This paper proposes a LLM-Augmenter system, which augments a black-box LLM with a set of plug-and-play modules.
  arXiv  Detail & Related papers  (2023-02-24T18:48:43Z)

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.