{"paper_id":"2402.18485","title":"A Multimodal Foundation Agent for Financial Trading: Tool-Augmented, Diversified, and Generalist","authors":["Wentao Zhang","Lingxuan Zhao","Haochong Xia","Shuo Sun","Jiaze Sun","Molei Qin","Xinyi Li","Yuqing Zhao","Yilei Zhao","Xinyu Cai","Longtao Zheng","Xinrun Wang","Bo An"],"paper_date":"2024-02-28","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2402.18485v3","page_url":"https://thequant.space/flowcharts/a-multimodal-foundation-agent-for-financial-trading-tool-au/","scores":{"empirical_rigor":8,"math_complexity":4,"quadrant":"Street Traders"},"proposes":{"status":"available","statement":"The paper proposes FinAgent, a multimodal foundational trading agent with tool augmentation that processes numerical, textual, and visual data for financial trading.","quote":"To address these challenges, we present FinAgent, a multimodal foundational agent with tool augmentation for financial trading.","summary":"Financial trading is a crucial component of the markets, informed by a multimodal information landscape encompassing news, prices, and Kline charts, and encompasses diverse tasks such as quantitative trading and high-frequency trading with various assets. While advanced AI techniques like deep…","strategy":"FinAgent"},"assumptions":{"status":"available","items":[{"assumption":"The study focuses on trading a single asset (stock or cryptocurrency) at a time.","type":"market","stated":"explicit","quote":"Specifically, we focus on single asset (e.g., stock or Crypto) trading."},{"assumption":"The reward function includes the change of market capital and accounts for commission fees.","type":"execution","stated":"explicit","quote":"The reward function leverages the change of market capitals (earned/lost money) [15] with consideration of commission fee [37, 44]."}]},"datasets":{"status":"available","items":[{"name":"Six real‑world datasets (five US stock markets and one cryptocurrency)","provider":null,"kind":"market-prices","access":"unknown","period":null,"universe":null,"quote":"To conduct a thorough evaluation of FinAgent, we evaluate it across 6 real‑world datasets. These included five datasets from the US stock markets and one is the cryptocurrency."},{"name":"Asset price data (open, high, low, close, adj close)","provider":null,"kind":"market-prices","access":"unknown","period":null,"universe":null,"quote":"i) Asset Price at the day‑level, including price data for open, high, low, close, and adj close;"},{"name":"Visual data (historical Kline and trading charts)","provider":null,"kind":"alternative","access":"unknown","period":null,"universe":null,"quote":"ii) Visual Data consists of historical Kline charts and trading charts, which are visual representations of asset market data and trading process on a daily basis;"},{"name":"Asset news coverage (Bloomberg Technology, Seeking Alpha, CNBC, etc.)","provider":null,"kind":"text-news","access":"unknown","period":null,"universe":null,"quote":"iii) Asset News coverage with daily updates from various esteemed sources, including Bloomberg Technology, Seeking Alpha, CNBC Television, and more,","known_id":"bloomberg"},{"name":"Expert Guidance (financial experts)","provider":null,"kind":"other","access":"unknown","period":null,"universe":null,"quote":"iv) Expert Guidance provided by financial experts as the auxiliary information,"}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"yes","headline_net_of_costs":null,"detail":"Commission fees are incorporated into the reward function as part of the trading cost model.","quote":"The reward function leverages the change of market capitals (earned/lost money) [15] with consideration of commission fee [37, 44].","source":"evidence_card","rule":"R2","candidates":{"front_matter":"unclear","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"Commission fees are incorporated into the reward function as part of the trading cost model.","card_quote":"The reward function leverages the change of market capitals (earned/lost money) [15] with consideration of commission fee [37, 44].","meaning":"costs are part of the reported headline results"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"FinAgent achieves over 36% average improvement on profit compared to 9 (or 12) state‑of‑the‑art baselines across six datasets.","kind":"performance","quote":"FinAgent significantly outperforms 9 state‑of‑the‑art baselines in terms of 6 financial metrics with over 36% average improvement on profit."},{"finding":"FinAgent attains a 92.27% return, representing an 84.39% relative improvement on one dataset.","kind":"performance","quote":"Specifically, a 92.27% return (a 84.39% relative improvement) is achieved on one dataset."},{"finding":"Ablation shows adding low‑level reflection increases ARR% by up to 101% and reduces risk by up to 44% for TSLA and ETHUSD.","kind":"performance","quote":"When compared to using solely M and ML, the integration of the low‑level reflection module leads to an impressive increase in ARR% by 45% to 101% for TSLA, and ETHUSD, and cutting risk by 14% to 44%."},{"finding":"Adding high‑level reflection further improves ARR% and Sharpe Ratio while slightly increasing maximum drawdown.","kind":"performance","quote":"When comparing the ML and MLH, the addition of the high‑level reflection module significantly enhances the ARR% and SR, while notably reducing risk. This improvement comes with a minor trade‑off: a slight 7% rise in MDD% for TSLA."}],"reported_performance":{"strategy":"FinAgent","sharpe":null,"sharpe_benchmark":null,"annual_return_pct":null,"total_return_pct":92.27,"max_drawdown_pct":null},"out_of_sample":{"method":"holdout","headline_out_of_sample":true}},"independent_evidence":{"status":"available","supports":[],"contradicts":[],"qualifies":[],"compared_with":39,"citations":{"cited_by_in_corpus":52,"by_independent_authors":42,"how_cited":{"used_as_baseline":2,"uses_method_or_data":7,"extends":0,"discusses_findings":0,"mentions":31,"unclassified":12},"independent_papers":[{"paper":{"paper_id":"2509.05080","title":"MM-DREX: Multimodal-Driven Dynamic Routing of LLM Experts for Financial Trading","paper_date":"2025-09-05","authors":["Yang Chen","Yueheng Jiang","Zhaozhao Ma","Yuchen Cao","Jacky Keung","Kun Kuang"],"empirical_rigor":6,"page_url":"https://thequant.space/flowcharts/mm-drex-multimodal-driven-dynamic-routing-of-llm-experts-fo/"},"how":"used_as_baseline","quote":null},{"paper":{"paper_id":"2411.08899","title":"FinVision: A Multi-Agent Framework for Stock Market Prediction","paper_date":"2024-10-29","authors":["Sorouralsadat Fatemi","Yuheng Hu"],"empirical_rigor":5,"page_url":"https://thequant.space/flowcharts/finvision-a-multi-agent-framework-for-stock-market-predicti/"},"how":"used_as_baseline","quote":null},{"paper":{"paper_id":"2508.06497","title":"Forecasting Commodity Price Shocks Using Temporal and Semantic Fusion of Prices Signals and Agentic Generative AI Extracted Economic News","paper_date":"2025-07-24","authors":["Mohammed-Khalil Ghali","Cecil Pang","Oscar Molina","Carlos Gershenson-Garcia","Daehan Won"],"empirical_rigor":8.5,"page_url":"https://thequant.space/flowcharts/forecasting-commodity-price-shocks-using-temporal-and-semant/"},"how":"uses_method_or_data","quote":null},{"paper":{"paper_id":"2502.18471","title":"FinBloom: Knowledge Grounding Large Language Model with Real-time Financial Data","paper_date":"2025-02-04","authors":["Ankur Sinha","Chaitanya Agarwal","Pekka Malo"],"empirical_rigor":8,"page_url":"https://thequant.space/flowcharts/finbloom-knowledge-grounding-large-language-model-with-real/"},"how":"uses_method_or_data","quote":null},{"paper":{"paper_id":"2505.07078","title":"Can LLM-based Financial Investing Strategies Outperform the Market in Long Run?","paper_date":"2025-05-11","authors":["Weixian Waylon Li","Hyeonjun Kim","Mihai Cucuringu","Tiejun Ma"],"empirical_rigor":8,"page_url":"https://thequant.space/flowcharts/can-llm-based-financial-investing-strategies-outperform-the/"},"how":"uses_method_or_data","quote":null},{"paper":{"paper_id":"2610.10407","title":"SOTA: Stock Options Trading Agents Guided by Option-Implied Return Distributions","paper_date":"2026-10-07","authors":["Yizhen Xie","Mengyang Liu"],"empirical_rigor":8,"page_url":"https://thequant.space/flowcharts/sota-stock-options-trading-agents-guided-by-option-implied-return-distributions/"},"how":"uses_method_or_data","quote":null},{"paper":{"paper_id":"2508.02366","title":"Language Model Guided Reinforcement Learning in Quantitative Trading","paper_date":"2025-08-04","authors":["Adam Darmanin","Vince Vella"],"empirical_rigor":6,"page_url":"https://thequant.space/flowcharts/language-model-guided-reinforcement-learning-in-quantitative/"},"how":"uses_method_or_data","quote":null},{"paper":{"paper_id":"2510.17393","title":"3S-Trader: A Multi-LLM Framework for Adaptive Stock Scoring, Strategy, and Selection in Portfolio Optimization","paper_date":"2025-10-20","authors":["Kefan Chen","Hussain Ahmad","Diksha Goel","Claudia Szabo"],"empirical_rigor":6,"page_url":"https://thequant.space/flowcharts/3s-trader-a-multi-llm-framework-for-adaptive-stock-scoring/"},"how":"uses_method_or_data","quote":null},{"paper":{"paper_id":"2408.06361","title":"Large Language Model Agent in Financial Trading: A Survey","paper_date":"2024-07-26","authors":["Han Ding","Yinheng Li","Junhao Wang","Hang Chen","Doudou Guo","Yunbai Zhang"],"empirical_rigor":5.5,"page_url":"https://thequant.space/flowcharts/large-language-model-agent-in-financial-trading-a-survey/"},"how":"uses_method_or_data","quote":null}],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":false,"code_url":null,"note":"No repository presented by the authors as their own was found in the paper. Absence here is not proof that no code exists."},"replication":{"status":"none_found","card":null,"note":"No replication of this paper has been run by thequant.space yet."},"extraction":{"text_source":"fulltext","model":"openai/gpt-oss-120b","version":1,"extracted_at":"2026-11-14T16:35:24Z"},"provenance":{"claims":"Everything here is what the authors state in the paper. Nothing was re-run or reproduced by the platform.","quotes":"Assumptions, datasets, cost detail and findings were extracted by a language model. Every quote is verbatim paper text. Where an item has `support`, \"full\" means every figure in it appears in the quoted sentences and \"partial\" means a figure is stated elsewhere in the paper; items with a figure found nowhere in the paper were discarded.","datasets_known":"datasets.known lists well-known vendors and public datasets by a stable id, found by a dictionary over the full text where the sentence says the data is used; it is independent of the model and covers only names in the dictionary.","flags":"transaction_costs.modeled, results.evaluation_type and results.out_of_sample.method are platform classifications (structured-decision models); see https://thequant.space/score-guide/#penalty-flags","comparisons":"supports / contradicts / qualifies are a language model's reading of two papers' quoted findings, restricted to closely related papers that share no author. They are pointers to read, not verified replications, and coverage is partial.","implementation":"code_url is a repository the paper itself presents as its own (rule-scored, ambiguous cases judged by a model), checked for liveness when it was recorded."},"docs":"https://thequant.space/agents/"}