{"paper_id":"2311.15548","title":"Deficiency of Large Language Models in Finance: An Empirical Examination of Hallucination","authors":["Haoqiang Kang","Xiao-Yang Liu"],"paper_date":"2023-11-27","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2311.15548v1","page_url":"https://thequant.space/flowcharts/deficiency-of-large-language-models-in-finance-an-empirical/","scores":{"empirical_rigor":6,"math_complexity":1,"quadrant":"Street Traders"},"proposes":{"status":"available","statement":"The paper proposes evaluating four practical methods—few-shot learning, Decoding by Contrasting Layers (DoLa), Retrieval Augmentation Generation (RAG), and prompt-based tool learning—to mitigate hallucinations in financial tasks.","quote":"we evaluate the efficacy of four practical methods, including few-shot learning, Decoding by Contrasting Layers (DoLa), the Retrieval Augmentation Generation (RAG) method and the prompt-based tool learning method","summary":"The hallucination issue is recognized as a fundamental deficiency of large language models (LLMs), especially when applied to fields such as finance, education, and law. Despite the growing concerns, there has been a lack of empirical investigation. In this paper, we provide an empirical…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The study focuses on factual inconsistency as the primary type of hallucination.","type":"other","stated":"explicit","quote":"In this work, we focus particularly on factual inconsistency, as it represents a serious and frequent type of error."}]},"datasets":{"status":"available","items":[{"name":"historical stock price queries","provider":null,"kind":"market-prices","access":"unknown","period":null,"universe":null,"quote":"assess LLM models’ capacity of querying historical stock prices"},{"name":"acronym recognition and terminology explanation tasks","provider":null,"kind":"benchmark","access":"unknown","period":null,"universe":null,"quote":"Table 1: Results of the acronym recognition task and terminology explanation task"}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"unclear"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R1","candidates":{"front_matter":"n/a","evidence_card":"no","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"forecast-accuracy","findings":[{"finding":"Off-the-shelf LLMs experience serious hallucination behaviors in financial tasks.","kind":"negative","quote":"our major finding is that off-the-shelf LLMs experience serious hallucination behaviors in financial tasks"},{"finding":"RAG significantly improves factuality in finance, raising accuracy and FactScore across models.","kind":"performance","quote":"RAG significantly improves the factuality in finance."},{"finding":"Few-shot prompting reduces errors but still leaves a noticeable gap in MAE and accuracy.","kind":"performance","quote":"while the utilization of few-shot prompting allows models to generate responses that are closer to correct answers, there still exists a gap in terms of MAE and accuracy"},{"finding":"GPT‑3.5‑turbo and GPT‑4 often abstain from answering stock‑symbol and stock‑price questions, avoiding potential misinformation.","kind":"performance","quote":"GPT3.5-turbo and GPT4 opt for a more cautious approach, abstaining from generating responses to some of stock symbol recognition questions and all stock-price-related questions"},{"finding":"Zero‑shot Llama2‑7B and Llama2‑7B‑chat exhibit alarmingly high MAE of 6357.6 USD and 6380.5 USD respectively when predicting stock prices.","kind":"negative","quote":"models such as Llama2-7B and Llama2-7B-chat exhibit alarmingly high Mean Absolute Errors (MAE) of 6357.6 USD and 6380.5 USD, respectively"}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":27,"citations":{"cited_by_in_corpus":4,"by_independent_authors":4,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":3,"unclassified":1},"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":"2026-11-14T18:18:38Z"},"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/"}