AI Paper Atlas

Research map / Large language models / Agents, code and math

LEXam: Benchmarking Legal Reasoning on 340 Law Exams

Yu Fan, Jingwei Ni, Jakob Merane, Yang Tian, Yoan Hermstrüwer, Yinya Huang, Mubashara Akhtar, Etienne Salimbeni, Florian Geering, Oliver Dreyer, Daniel Brunner, Markus Leippold and 5 more

ICLR 2026 · Large language models › Agents, code and math · approach: Mathematical and structural reasoning benchmarks

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
increase number of answer choices from 4 to 32 on perturbed MCQs (ablation)Gemini-2.5-Pro · LEXam perturbed MCQsaccuracy: 68.61 -> 35.62Hurt
increase number of answer choices from 4 to 32 (ablation)DeepSeek-V3 · LEXam perturbed MCQsaccuracy: 58.57 -> 16.03Hurt
increase number of answer choices from 4 to 32 (ablation)GPT-4o · LEXam perturbed MCQsaccuracy: 53.73 -> 21.81Hurt
use reasoning models instead of non-reasoning modelsLLMs · LEXam open questionsjudge score: 70.20 (GPT-5) vs 57.50 (GPT-4.1)Improved
phrase MCQ negativelyreasoning models · LEXam MCQ-16accuracyHurt
increase number of statements from 4 to 5reasoning models · LEXam MCQ-16accuracyNo clear effect
evaluate German instead of English questionsLLMs · LEXam open questionsjudge scoreHurt
use minimum-score ensemble of GPT-4o, Qwen3-32B, DeepSeek-V3 as judgeLLM judge · Alt-test vs 3 legal expertswinning rate: 1.00Improved
add MCQ answer option 'None of the statements'LLMs · LEXam MCQsaccuracyHurt
include longer model answersLLMs · LEXam open questionsjudge scoreHurt
increase question lengthLLMs · LEXam open questionsjudge scoreNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy