SteuerLLM: Local specialized large language model for German tax law analysis
Large language models (LLMs) demonstrate strong general reasoning and language understanding, yet their performance degrades in domains governed by strict formal rules, precise terminology, and legally binding structure. Tax law exemplifies these challenges, as correct answers require exact statutory citation, structured legal argumentation, and numerical accuracy under rigid grading schemes. We algorithmically generate SteuerEx, the first open benchmark derived from authentic German university tax law examinations. SteuerEx comprises 115 expert-validated examination questions spanning six core tax law domains and multiple academic levels, and employs a statement-level, partial-credit evaluation framework that closely mirrors real examination practice. We further present SteuerLLM, a domain-adapted LLM for German tax law trained on a large-scale synthetic dataset generated from authentic examination material using a controlled retrieval-augmented pipeline. SteuerLLM (28B parameters) consistently outperforms general-purpose instruction-tuned models of comparable size and, in several cases, substantially larger systems, demonstrating that domain-specific data and architectural adaptation are more decisive than parameter scale for performance on realistic legal reasoning tasks. All benchmark data, training datasets, model weights, and evaluation code are released openly to support reproducible research in domain-specific legal artificial intelligence. A web-based demo of SteuerLLM is available at https://steuerllm.i5.ai.fau.de.
Code (0)
등록된 구현이 없습니다.
Tasks
Legal ReasoningSimilar Papers 제목 키워드 기반
Can Continual Pre-training Bridge the Performance Gap between General-purpose and Specialized Language Models in the Medical Domain?
This paper narrows the performance gap between small, specialized models and significantly larger general-purpose models through domain adaptation via continual pre-training and merging. We address the scarcity of specia…
Domain AdaptationDomain-Adaptation through Synthetic Data: Fine-Tuning Large Language Models for German Law
Large language models (LLMs) often struggle in specialized domains such as legal reasoning due to limited expert knowledge, resulting in factually incorrect outputs or hallucinations. This paper presents an effective met…
parameter-efficient fine-tuningSynthetic Data GenerationQuestion AnsweringLegal ReasoningG-SciEdBERT: A Contextualized LLM for Science Assessment Tasks in German
The advancement of natural language processing has paved the way for automated scoring systems in various languages, such as German (e.g., German BERT [G-BERT]). Automatically scoring written responses to science questio…
Language ModellingLarge Language ModelDETECT: Determining Ease and Textual Clarity of German Text Simplifications
Current evaluation of German automatic text simplification (ATS) relies on general-purpose metrics such as SARI, BLEU, and BERTScore, which insufficiently capture simplification quality in terms of simplicity, meaning pr…
Text Simplification2nd Swiss German Speech to Standard German Text Shared Task at SwissText 2022
We present the results and findings of the 2nd Swiss German speech to Standard German text shared task at SwissText 2022. Participants were asked to build a sentence-level Swiss German speech to Standard German text syst…
Sentence