{"paper_id":"2508.20097","title":"Can LLMs Identify Tax Abuse?","authors":["Andrew Blair-Stanek","Nils Holzenberger","Benjamin Van Durme"],"paper_date":"2025-08-10","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2508.20097v1","page_url":"https://thequant.space/flowcharts/can-llms-identify-tax-abuse/","scores":{"empirical_rigor":7.5,"math_complexity":2,"quadrant":"Street Traders"},"proposes":{"status":"available","statement":"The paper introduces the Shelter Check dataset of 36 U.S. tax‑minimization strategies and benchmarks several leading LLMs on interpreting, verifying, and generating such strategies.","quote":"We introduce a new domain-expert crafted dataset called Shelter Check, derived from public sources that describe schemes used in the past by taxpayers to minimize taxes.","summary":"We investigate whether large language models can discover and analyze U.S. tax-minimization strategies. This real-world domain challenges even seasoned human experts, and progress can reduce tax revenue lost from well-advised, wealthy taxpayers. We evaluate the most advanced LLMs on their ability…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The tax‑minimization strategies in the study depend on textual tax‑law authorities rather than on asset‑valuation uncertainty.","type":"data","stated":"implied","quote":"3 The Shelter Check Dataset The Shelter Check dataset currently contains 36 known U.S. tax-minimization strategies that depend upon applying tex-tual authorities. Why only 36?","support":"full"}]},"datasets":{"status":"available","items":[{"name":"Shelter Check","provider":"public sources (IRS listed transactions, court decisions)","kind":"text-news","access":"public","period":null,"universe":null,"quote":"We introduce a new domain-expert crafted dataset called Shelter Check, derived from public sources that describe schemes used in the past by taxpayers to minimize taxes. The dataset includes the text of the tax-law authorities, like tax code sections or court decisions, that enabled the scheme.","support":"full"}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"none"},"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":"unclear","findings":[{"finding":"LLM‑based reasoning identified an entirely novel tax strategy, demonstrating potential for tax agencies.","kind":"other","quote":"Notably, LLM-based reasoning identified an entirely novel tax strategy, highlighting these models' potential to revolutionize tax agencies' fight against tax abuse.","support":"full"},{"finding":"All three reasoning models (o3, claude‑4, gemini‑2.5) outperformed the baseline llama‑3.3‑70B on the step‑cloze and generation tasks.","kind":"performance","quote":"The three reasoning models outperformed the baseline llama-3.3-70B. so that replacing a specified correct step with the adversarial step would keep the strategy from working (i.e. achieving all listed goals).","support":"partial"},{"finding":"When adversarial steps were added, the reasoning models initially showed agreement bias, but after iterative refinement they returned 0/36 correct strategies, indicating resistance to finding non‑viable strategies.","kind":"negative","quote":"But the LLMs proved remarkably resistant to finding that a strategy does not work. We improved the adversarial steps and expressly eliminated unwarranted factual assumptions to minimize the number of strategies the models found cor-rect, until all three reasoning models returned 0/36 as being correct.","support":"full"}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":8,"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":true,"code_url":"https://github.com/BlairStanek/LLM-Tax-Abuse","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":2,"extracted_at":"2026-11-14T19:48:32Z"},"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/"}