{"paper_id":"2510.19203","title":"Aligning Multilingual News for Stock Return Prediction","authors":["Yuntao Wu","Lynn Tao","Ing-Haw Cheng","Charles Martineau","Yoshio Nozawa","John Hull","Andreas Veneris"],"paper_date":"2025-10-22","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2510.19203v1","page_url":"https://thequant.space/flowcharts/aligning-multilingual-news-for-stock-return-prediction/","scores":{"empirical_rigor":6,"math_complexity":7,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a sentence‑level alignment method for multilingual news using optimal transport to improve stock return prediction.","quote":"We propose a method to align sentences in multilingual news articles using optimal transport, identifying semantically similar content across languages.","summary":"News spreads rapidly across languages and regions, but translations may lose subtle nuances. We propose a method to align sentences in multilingual news articles using optimal transport, identifying semantically similar content across languages. We apply this method to align more than 140,000 pairs…","strategy":"aligned sentence embeddings long-short portfolio"},"assumptions":{"status":"available","items":[{"assumption":"Articles published within 30 minutes before market open are linked to that day’s stock price and return.","type":"data","stated":"explicit","quote":"articles published between 30 minutes before market open on day"},{"assumption":"Only articles with a ticker score ≥75 are considered, implying medium‑to‑high relevance to the stock.","type":"data","stated":"explicit","quote":"we include English and foreign-language articles that are explicitly associated with a single stock ticker with a ticker score ≥75, indicating medium-to-high relevance to that stock."},{"assumption":"Sentences in each article are assigned equal probability mass in the optimal transport formulation.","type":"model","stated":"explicit","quote":"Sentences are assigned equal probability mass, i.e., uniformly distributed."},{"assumption":"Cost matrix is defined by pairwise cosine distance scaled to [0,1] with min‑max normalization.","type":"statistical","stated":"explicit","quote":"The cost matrix is defined using pairwise cosine distance"},{"assumption":"Trading signals are formed from daily long‑short portfolios without accounting for transaction costs.","type":"execution","stated":"implied","quote":"Portfolios constructed from aligned embeddings exhibit even higher Sharpe ratios for both Japanese and English texts, suggesting that sentences capturing common themes across languages provide clearer signals of stock performance."}]},"datasets":{"status":"available","items":[{"name":"Bloomberg news feed","provider":"Bloomberg","kind":"text-news","access":"commercial","period":"2008–2024","universe":"≈3,500 Tokyo‑listed stocks, English and Japanese articles","quote":"We collect global news articles from the Bloomberg terminal news feed covering the period 2008–2024.","known_id":"bloomberg"},{"name":"Compustat Global via WRDS","provider":"Compustat Global (WRDS)","kind":"market-prices","access":"commercial","period":"2008–2024","universe":"Tokyo Stock Exchange stocks","quote":"Stock return data are obtained from Compustat Global using WRDS for the period 2008–2024.","known_id":"compustat"}],"known":[{"id":"bloomberg","name":"Bloomberg","provider":"Bloomberg L.P.","kind":"market-prices","access":"commercial","quote":"3 Data We collect global news articles from the Bloomberg terminal news feed covering the period 2008–2024.1 For each unique story identi-fier, we retain only the final update published within 24 hours of its initial appearance on Bloomberg.","mentions":6,"used_in":1},{"id":"compustat","name":"Compustat","provider":"S&P Global","kind":"fundamentals","access":"commercial","quote":"Stock return data are obtained from Compustat Global using WRDS for the period 2008–2024.","mentions":1,"used_in":1},{"id":"wrds","name":"WRDS","provider":"Wharton Research Data Services","kind":"other","access":"commercial","quote":"Stock return data are obtained from Compustat Global using WRDS for the period 2008–2024.","mentions":1,"used_in":1}],"sample":{"start":"2012","end":"2024","market":"Tokyo exchange 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":"Long‑short strategies based on aligned sentence embeddings achieve higher Sharpe ratios (e.g., 4.36 annualized Sharpe for Japanese aligned embeddings) than those using unaligned or full‑text embeddings.","kind":"performance","quote":"Portfolios constructed from aligned embeddings exhibit even higher Sharpe ratios for both Japanese and English texts, suggesting that sentences capturing common themes across languages provide clearer signals of stock performance."},{"finding":"Return scores built from aligned sentences show stronger correlations with realized stock returns than scores from unaligned or full‑text embeddings.","kind":"predictive-accuracy","quote":"Return scores constructed from aligned sentences show stronger correlations with realized stock returns."},{"finding":"Aligned sentences are sparser, more interpretable, and exhibit higher semantic similarity than those obtained via pure cosine similarity.","kind":"statistical","quote":"Aligned sentences are sparser, more interpretable, and exhibit higher semantic similarity."},{"finding":"In the Japanese market, full embeddings of Japanese news generate higher Sharpe than English news, indicating stronger predictive power of native‑language coverage.","kind":"performance","quote":"In the Japanese market, the full embedding of Japanese news generates higher Sharpe than English news."}],"reported_performance":{"strategy":"aligned sentence embeddings long-short portfolio","sharpe":4.36,"sharpe_benchmark":null,"annual_return_pct":null,"total_return_pct":null,"max_drawdown_pct":null},"out_of_sample":{"method":"walk-forward","headline_out_of_sample":true}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":22,"citations":{"cited_by_in_corpus":0,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":0},"independent_papers":[],"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":"2026-11-14T17:23:47Z"},"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/"}