{"paper_id":"2609.35086","title":"Retrieval-Augmented Diffusion Modeling for Stochastic Discount Factor Portfolios","authors":["Kelvin J. L. Koa","Xinyang Li","Ke-Wei Huang"],"paper_date":"2026-09-28","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2609.35086","page_url":"https://thequant.space/flowcharts/retrieval-augmented-diffusion-modeling-for-stochastic-discount-factor-portfolios/","scores":{"empirical_rigor":6,"math_complexity":7.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes RADAR, a retrieval-augmented diffusion framework that learns context‑aware market representations for SDF‑based portfolio optimization.","quote":"RADAR, a retrieval-augmented diffusion framework that learns context-aware market representations by conditioning on similar historical regimes to denoise multimodal features.","summary":"In this work, we study portfolio optimization under the stochastic discount factor (SDF) framework by learning market state representations that capture the underlying risk structures of financial data. This is challenging due to several factors: financial markets exhibit non-stationary dynamics…","strategy":"RADAR"},"assumptions":{"status":"available","items":[{"assumption":"Financial markets exhibit non‑stationary dynamics with shifting regimes.","type":"market","stated":"explicit","quote":"financial markets exhibit non-stationary dynamics with shifting regimes"},{"assumption":"Multimodal inputs such as price and news data contain stochastic noise.","type":"data","stated":"explicit","quote":"multimodal inputs such as price and news data often contain stochastic noise"},{"assumption":"Standard diffusion models assume isotropic Gaussian noise, which does not capture state‑dependent uncertainty.","type":"model","stated":"explicit","quote":"existing diffusion‑based approaches, while effective for modeling stochastic dynamics, rely on assumptions such as isotropic Gaussian noise that fail to capture the state‑dependent nature of financial uncertainty"},{"assumption":"A rolling‑window protocol with 4‑year training windows avoids look‑ahead bias.","type":"execution","stated":"explicit","quote":"we adopt a rolling‑window protocol: each window uses 4 years of data for training (with a 90:10 train/validation split) and the subsequent 1 year for out‑of‑sample testing, rolling forward by 1 year"},{"assumption":"The diffusion noise distribution can be conditioned on retrieved historical contexts.","type":"model","stated":"explicit","quote":"We introduce a state‑dependent diffusion formulation where the noise distribution is conditioned on retrieved historical contexts"}]},"datasets":{"status":"available","items":[{"name":"S&P 500 constituent stocks dataset","provider":"Yahoo Finance and FNSPID","kind":"market-prices","access":"public","period":"June 2011 to June 2020","universe":"377 U.S. equities that appear in both the S&P 500 index and the news corpus","quote":"We collect daily adjusted close prices from Yahoo Finance and financial news articles from FNSPID [10], and the final dataset includes 377 U.S. equities that appear in both the S&P 500 index and the news corpus. The duration spans from June 2011 to June 2020, covering approximately 2,270 trading days.","known_id":"yahoo-finance"}],"known":[{"id":"fnspid","name":"FNSPID","provider":"Dong et al.","kind":"text-news","access":"public","quote":"We collect daily adjusted close prices from Yahoo Finance and financial news articles from FNSPID [10], and the final dataset includes 377 U.S. equities that appear in both the S&P 500 index and the news corpus.","mentions":3,"used_in":1},{"id":"yahoo-finance","name":"Yahoo Finance","provider":"Yahoo","kind":"market-prices","access":"public","quote":"We collect daily adjusted close prices from Yahoo Finance and financial news articles from FNSPID [10], and the final dataset includes 377 U.S. equities that appear in both the S&P 500 index and the news corpus.","mentions":1,"used_in":1}],"sample":{"start":"2011-06","end":"2020-06","market":"S&P 500 constituent stocks","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":"RADAR achieves state‑of‑the‑art performance on key risk‑adjusted metrics such as Sharpe, Sortino, and Calmar ratios.","kind":"performance","quote":"RADAR consistently achieves state‑of‑the‑art performance across key risk‑adjusted metrics, including Sharpe, Sortino, and Calmar ratios."},{"finding":"Quintile portfolios formed from higher‑scoring assets consistently outperform lower‑scoring ones, indicating economically meaningful representations.","kind":"performance","quote":"As shown, the NAV curves exhibit a clear separation, with Q1 consistently outperforming the lower quintiles. This indicates that the RADAR embeddings capture economically meaningful cross‑sectional asset returns signals that translate into improved portfolio performance."},{"finding":"RADAR consistently attains the highest Sharpe ratio and cumulative return across various input lengths and rebalancing horizons.","kind":"performance","quote":"RADAR consistently achieves the highest Sharpe ratio and cumulative return across all six configurations."},{"finding":"The method generalizes well to multiple non‑financial domains, often matching or exceeding baselines.","kind":"performance","quote":"Overall, we find that RADAR is largely generalizable across multiple domains."}],"reported_performance":{"strategy":"RADAR","sharpe":1.062,"sharpe_benchmark":0.513,"annual_return_pct":28.5,"total_return_pct":250.2,"max_drawdown_pct":30.8},"out_of_sample":{"method":"walk-forward","headline_out_of_sample":true}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":48,"citations":{"cited_by_in_corpus":1,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":1},"independent_papers":[],"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/xinyangli5579-star/RADAR","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":"2026-11-14T17:04:15Z"},"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/"}