{"paper_id":"2405.00522","title":"DAM: A Universal Dual Attention Mechanism for Multimodal Timeseries Cryptocurrency Trend Forecasting","authors":["Yihang Fu","Mingyu Zhou","Luyao Zhang"],"paper_date":"2024-05-01","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2405.00522v1","page_url":"https://thequant.space/flowcharts/dam-a-universal-dual-attention-mechanism-for-multimodal-tim/","scores":{"empirical_rigor":5,"math_complexity":7,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a Dual Attention Mechanism (DAM) for multimodal cryptocurrency time‑series forecasting that fuses financial and sentiment data.","quote":"We designed DAM, a universal Dual Attention Mechanism for multimodal financial time-series data fusion.","summary":"In the distributed systems landscape, Blockchain has catalyzed the rise of cryptocurrencies, merging enhanced security and decentralization with significant investment opportunities. Despite their potential, current research on cryptocurrency trend forecasting often falls short by simplistically…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The target variable for forecasting is the daily closing price of Bitcoin.","type":"data","stated":"explicit","quote":"The target variable is the daily closing price."},{"assumption":"The last 70 days of data are used as a validation set.","type":"data","stated":"explicit","quote":"We split the last 70‑day data as the validation data."},{"assumption":"Sentiment from news and social media influences cryptocurrency price movements.","type":"market","stated":"implied","quote":"news or social media information on other stocks or cryptocurrencies could cause the co‑movements of the cryptocurrency."},{"assumption":"Variables with high correlation (high, low, open price, volume) are relevant for the model.","type":"statistical","stated":"implied","quote":"the variables high, low, open price, and volume show a high correlation with t…"}]},"datasets":{"status":"available","items":[{"name":"Bitcoin price data","provider":"CryptoCompare API","kind":"market-prices","access":"public","period":"2020-01-01 to 2023-06-28","universe":"Bitcoin","quote":"We queried cryptocurrency data (Bitcoin) ranges on a daily level from 2020-01-01 to 2023-06-28 by using CryptoCompare API."},{"name":"Bitcoin news sentiment","provider":"Nasdaq","kind":"text-news","access":"public","period":null,"universe":"Bitcoin","quote":"The crypto news information is queried from Nasdaq."},{"name":"Bitcoin social media posts","provider":"Kaggle public dataset","kind":"alternative","access":"public","period":null,"universe":"Bitcoin","quote":"The social media posts are collected from Kaggle public dataset.","known_id":"kaggle"}],"known":[{"id":"kaggle","name":"Kaggle","provider":"Kaggle","kind":"other","access":"public","quote":"We sourced Bitcoin data via the Cryptocompare API, extracted Bitcoin news sentiment from Nasdaq, and social media data from a public Kaggle dataset.","mentions":2,"used_in":2}],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"daily"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R3","candidates":{"front_matter":"n/a","evidence_card":"n/a","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"forecast-accuracy","findings":[{"finding":"The Dual Attention Mechanism improves prediction accuracy by up to 20 % compared with conventional models.","kind":"performance","quote":"Our dual attention mechanism increases a 20 percent improvement."},{"finding":"DAM‑LSTM outperforms NeuralProphet and other baselines, showing a 20 % improvement over simple concatenation fusion.","kind":"predictive-accuracy","quote":"Our Dual Attention Mechanism‑LSTM (DAM‑LSTM) model outperforms the NeuralProphet model and other selected models and shows a 20 percent improvement on the same LSTM architecture but with the simple concatenation fusion."},{"finding":"Integrating sentiment data raises TFT model prediction performance by about 20 % and LSTM model by about 5 %.","kind":"predictive-accuracy","quote":"The application of multimodal data confers an elevation of prediction by approximately 20 percent for the TFT model and 5 percent for the LSTM model."},{"finding":"Stationary data input improves the LSTM‑based model by nearly 50 % in MAE.","kind":"statistical","quote":"the stationary data input improves the LSTM‑based model by nearly 50% percent."},{"finding":"DAM‑LSTM achieves MAE 719.82 and MAPE 0.0297 on the multimodal test set.","kind":"performance","quote":"DAM‑LSTM 719.82 431.86 0.0297 Yes"}],"reported_performance":null,"out_of_sample":{"method":"holdout","headline_out_of_sample":null}},"independent_evidence":{"status":"available","supports":[],"contradicts":[],"qualifies":[{"paper":{"paper_id":"2412.06794","title":"Understanding the Impact of News Articles on the Movement of Market Index: A Case on Nifty 50","paper_date":"2024-11-22","authors":["Subhasis Dasgupta","Pratik Satpati","Ishika Choudhary","Jaydip Sen"],"empirical_rigor":8,"page_url":"https://thequant.space/flowcharts/understanding-the-impact-of-news-articles-on-the-movement-of/"},"question":"Does adding sentiment data improve prediction accuracy for financial time series?","this_paper_finding":"Integrating sentiment data raises TFT model prediction performance by about 20 % and LSTM model by about 5 %.","other_paper_finding":"Using VADER sentiment scores with up to 3 lags marginally improves RMSE from 134.85 to 134.15 compared to a model without sentiment.","other_paper_quote":"The RMSE observed was 134.85. That way, with VADER sentiment scores on the news data, a marginal improvement in RMSE was observed with up to lag 3 data.","note":"A reports ~20% gain for TFT and ~5% for LSTM; B shows RMSE drop from 134.85 to 134.15 (~0.5%). Both find sentiment beneficial, though magnitude and metric differ.","confidence":0.738}],"compared_with":40,"citations":{"cited_by_in_corpus":2,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":2},"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":"2026-11-14T18:10:13Z"},"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/"}