{"paper_id":"2403.12285","title":"FinLlama: Financial Sentiment Classification for Algorithmic Trading Applications","authors":["Thanos Konstantinidis","Giorgos Iacovides","Mingxue Xu","Tony G. Constantinides","Danilo Mandic"],"paper_date":"2024-03-18","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2403.12285v1","page_url":"https://thequant.space/flowcharts/finllama-financial-sentiment-classification-for-algorithmic/","scores":{"empirical_rigor":6,"math_complexity":3.5,"quadrant":"Street Traders"},"proposes":{"status":"available","statement":"The paper proposes FinLlama, a fine‑tuned Llama 2 7B model for financial sentiment classification and strength quantification.","quote":"Our proposed FinLlama is one such solution, which is obtained by fine-tuning a pre-trained LLM (namely Llama 2 7B [2]) on specialised, labelled and publicly available financial news datasets.","summary":"There are multiple sources of financial news online which influence market movements and trader’s decisions. This highlights the need for accurate sentiment analysis, in addition to having appropriate algorithmic trading techniques, to arrive at better informed trading decisions. Standard lexicon…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"A risk‑free rate of 0% is used for Sharpe ratio calculations.","type":"market","stated":"explicit","quote":"a 0% rate is commonly used and is adopted in our analysis."}]},"datasets":{"status":"available","items":[{"name":"Four labelled publicly available financial news datasets","provider":null,"kind":"text-news","access":"public","period":null,"universe":null,"quote":"Our training data was a combination of four labelled publicly available financial news datasets, resulting in a comprehensive collection of 34,180 labelled samples."},{"name":"Financial news articles","provider":null,"kind":"text-news","access":"public","period":"2015-2021","universe":null,"quote":"204,017 articles dating between 2015 to 2021 were collected from online sources."},{"name":"Yahoo Finance S&P 500 daily returns","provider":"Yahoo Finance","kind":"market-prices","access":"public","period":"2015-2021","universe":"S&P 500 constituents","quote":"Financial market data were collected for the same time period from Yahoo Finance. These collected market data contained daily stock returns for the 500 companies in our Investable Universe (S&P 500), resulting in 1,672 days of stock returns data for each company.","known_id":"yahoo-finance"}],"known":[{"id":"bloomberg","name":"Bloomberg","provider":"Bloomberg L.P.","kind":"market-prices","access":"commercial","quote":"To the best of our knowledge, BloombergGPT [15] is the only pre-trained finance-specific LLM, as Bloomberg was able to train the model using data collected over a span of 40 years.","mentions":1,"used_in":1}],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"no","headline_net_of_costs":null,"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":"FinLlama’s 35% long‑short portfolio achieved the highest cumulative return of 308.2%, outperforming all other sentiment methods.","kind":"performance","quote":"Cumulative Returns (%) 204.6 100.4 130.6 213.0 308.2 83.1"},{"finding":"FinLlama generated significantly higher returns while reducing portfolio risk, as shown by higher Sharpe and lower annualized volatility.","kind":"performance","quote":"Overall, our FinLlama model successfully generated significantly higher returns for investors, whilst simultaneously reducing portfolio risk, as indicated by the higher Sharpe ratio and lower annualized volatility."},{"finding":"Deep learning approaches, including FinLlama, outperformed lexicon‑based methods in cumulative returns, especially during high‑volatility periods.","kind":"performance","quote":"Observe that the deep learning approaches outperformed the lexicon-based approaches in terms of cumulative returns, particularly those relying on general-purpose dictionaries (HIV-4 and VADER)."}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"available","supports":[],"contradicts":[],"qualifies":[],"compared_with":22,"citations":{"cited_by_in_corpus":6,"by_independent_authors":6,"how_cited":{"used_as_baseline":0,"uses_method_or_data":3,"extends":0,"discusses_findings":0,"mentions":1,"unclassified":2},"independent_papers":[{"paper":{"paper_id":"2606.08285","title":"Execution Realism and Reproducibility in LLM-Based Trading Systems: A Systematic Scoping Review and Evidence Audit","paper_date":"2026-06-06","authors":["Junyi Yao","Zihao Zheng","Baichuan Li","Jiayu Long"],"empirical_rigor":8,"page_url":"https://thequant.space/flowcharts/execution-realism-and-reproducibility-in-llm-based-trading-systems-a-systematic/"},"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},{"paper":{"paper_id":"2503.21422","title":"From Deep Learning to LLMs: A survey of AI in Quantitative Investment","paper_date":"2025-03-27","authors":["Bokai Cao","Saizhuo Wang","Xinyi Lin","Xiaojun Wu","Haohan Zhang","Lionel M. Ni"],"empirical_rigor":1.5,"page_url":"https://thequant.space/flowcharts/from-deep-learning-to-llms-a-survey-of-ai-in-quantitative-i/"},"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":"2026-11-14T18:13:04Z"},"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/"}