{"paper_id":"2409.06289","title":"Automate Strategy Finding with LLM in Quant Investment","authors":["Zhizhuo Kou","Holam Yu","Junyu Luo","Jingshu Peng","Xujia Li","Chengzhong Liu","Juntao Dai","Lei Chen","Sirui Han","Yike Guo"],"paper_date":"2024-09-10","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2409.06289v4","page_url":"https://thequant.space/flowcharts/automate-strategy-finding-with-llm-in-quant-investment/","scores":{"empirical_rigor":6,"math_complexity":4.5,"quadrant":"Street Traders"},"proposes":{"status":"available","statement":"The paper proposes a novel three‑stage framework that uses Large Language Models within a risk‑aware multi‑agent system to automatically generate and evaluate alpha factor candidates for quantitative trading.","quote":"We present a novel three-stage framework leveraging Large Language Models (LLMs) within a risk-aware multi-agent system for automate strategy finding in quantitative finance.","summary":"We present a novel three-stage framework leveraging Large Language Models (LLMs) within a risk-aware multi-agent system for automate strategy finding in quantitative finance. Our approach addresses the brittleness of traditional deep learning models in financial applications by: employing…","strategy":"LLM-driven framework"},"assumptions":{"status":"available","items":[{"assumption":"The multi‑agent evaluation methodology presupposes persistent historical relationships between market conditions and alpha performance.","type":"market","stated":"explicit","quote":"our multi-agent evaluation methodology presupposes persistent historical relationships between market conditions and alpha performance, an assumption that may prove tenuous during market regime shifts."},{"assumption":"System efficacy depends on the quality of input documents.","type":"data","stated":"explicit","quote":"system efficacy is contingent upon input document quality, potentially perpetuating inherent biases"},{"assumption":"Alpha factors are evaluated using the Information Coefficient (IC) as a measure of predictive power.","type":"statistical","stated":"explicit","quote":"alpha factor predictive power is measured using the Information Coefficient (IC) 1."},{"assumption":"Backtesting uses historical market data without explicit modeling of slippage or market impact.","type":"execution","stated":"implied","quote":"Selected alpha factors undergo rigorous backtesting using historical market data, with key evaluation metrics including the IC"}]},"datasets":{"status":"available","items":[{"name":"SSE 50 Index constituents","provider":null,"kind":"market-prices","access":"public","period":"Jan 2023 to Jan 2024","universe":"50 constituent companies of the SSE 50 Index","quote":"overall framework significantly outperforms all benchmarks with 53.17% cumulative return on SSE50 (Jan 2023 to Jan 2024)"},{"name":"CSI300 constituents","provider":"Tushare API","kind":"market-prices","access":"public","period":"Jan 2019 to Dec 2023","universe":"300 largest companies listed on China’s Shanghai and Shenzhen exchanges","quote":"CSI300 constituents, particularly in H1 2023 (192.27% annual return vs. benchmark 9.13%)","known_id":"tushare"},{"name":"SP500 constituents","provider":"CRSP platform","kind":"market-prices","access":"commercial","period":"Jan 2019 to Dec 2023","universe":"500 large U.S. companies traded on American stock exchanges","quote":"SP500 constituents (93.61% in H1 2021, 118.24% in H1 2023)","known_id":"crsp"},{"name":"Financial reports (quarterly and annual) of index constituent companies","provider":null,"kind":"fundamentals","access":"public","period":"Jan 2019‑Jun 2024","universe":"SSE 50, CSI300, SP500 constituents","quote":"Financial Reports Quarterly and Annual reports from Index constituent companies"}],"known":[],"sample":{"start":"2023-01","end":"2024-01","market":"SSE50","asset_class":"equities","frequency":"daily"},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"no","headline_net_of_costs":false,"detail":null,"quote":null,"source":"front_matter","rule":"R1","candidates":{"front_matter":"no","evidence_card":"unclear","evidence_card_quoted":false},"meaning":"a trading or execution result is reported without costs"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"The framework achieved a 53.17% cumulative return on the SSE50 index from Jan 2023 to Jan 2024, outperforming all benchmarks.","kind":"performance","quote":"overall framework significantly outperforms all benchmarks with 53.17% cumulative return on SSE50 (Jan 2023 to Jan 2024)"},{"finding":"Out‑of‑sample evaluation yielded an Information Coefficient of 0.047 and a Sharpe Ratio of 1.73.","kind":"performance","quote":"the complete model achieved superior performance, with an out-of-sample IC of 0.047 and Sharpe Ratio of 1.73"},{"finding":"In the Chinese A‑Share market, the strategy generated a 192.27% annual return for CSI300 constituents in H1 2023 versus a 9.13% benchmark.","kind":"performance","quote":"192.27% annual return vs. benchmark 9.13%"},{"finding":"During the H1 2022 market downturn, the strategy maintained positive returns (12.78% in CSI300, 2.77% in SP500) while benchmarks fell sharply.","kind":"performance","quote":"maintained positive returns in both markets (12.78% in CSI300, 2.77% in SP500) while their respective benchmarks declined significantly (-30.37% and -44.22%)"}],"reported_performance":{"strategy":"LLM-driven framework","sharpe":0.287,"sharpe_benchmark":-0.063,"annual_return_pct":53.17,"total_return_pct":53.17,"max_drawdown_pct":null},"out_of_sample":{"method":"holdout","headline_out_of_sample":true}},"independent_evidence":{"status":"available","supports":[],"contradicts":[],"qualifies":[],"compared_with":21,"citations":{"cited_by_in_corpus":11,"by_independent_authors":10,"how_cited":{"used_as_baseline":1,"uses_method_or_data":3,"extends":0,"discusses_findings":0,"mentions":5,"unclassified":2},"independent_papers":[{"paper":{"paper_id":"2505.15155","title":"R&D-Agent-Quant: A Multi-Agent Framework for Data-Centric Factors and Model Joint Optimization","paper_date":"2025-05-21","authors":["Yuante Li","Xu Yang","Xiao Yang","Minrui Xu","Xisen Wang","Weiqing Liu"],"empirical_rigor":8,"page_url":"https://thequant.space/flowcharts/rd-agent-quant-a-multi-agent-framework-for-data-centric-fa/"},"how":"used_as_baseline","quote":null},{"paper":{"paper_id":"2509.01393","title":"Adaptive Alpha Weighting with PPO: Enhancing Prompt-Based LLM-Generated Alphas in Quant Trading","paper_date":"2025-09-01","authors":["Qizhao Chen","Hiroaki Kawashima"],"empirical_rigor":8,"page_url":"https://thequant.space/flowcharts/adaptive-alpha-weighting-with-ppo-enhancing-prompt-based-ll/"},"how":"uses_method_or_data","quote":null},{"paper":{"paper_id":"2610.14959","title":"AlphaRLM: Aligning Language Models for Literature-Grounded Symbolic Factor Discovery","paper_date":"2026-10-10","authors":["Jie Huang","Yuecheng Hong","Yao Chen","Hao Zhong","Linfeng Xiao","Xudong Liu"],"empirical_rigor":8,"page_url":"https://thequant.space/flowcharts/alpharlm-aligning-language-models-for-literature-grounded-symbolic-factor/"},"how":"uses_method_or_data","quote":null},{"paper":{"paper_id":"2507.08584","title":"To Trade or Not to Trade: An Agentic Approach to Estimating Market Risk Improves Trading Decisions","paper_date":"2025-07-11","authors":["Dimitrios Emmanoulopoulos","Ollie Olby","Justin Lyon","Namid R. Stillman"],"empirical_rigor":6,"page_url":"https://thequant.space/flowcharts/to-trade-or-not-to-trade-an-agentic-approach-to-estimating/"},"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":true,"code_url":"https://github.com/kouzhizhuo/Automate-Strategy-Finding-with-LLM-in-Quant-investment","note":null},"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":"2026-11-14T18:02:40Z"},"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/"}