LetTokenBurn
Sign in
Verified100%Harbor adapterKnowledge & Reasoning

Case HOLD

Case Holdings On Legal Decisions

CaseHOLD (Case Holdings On Legal Decisions) is a multiple choice question answering scenario where the task is to identify the relevant holding of a cited case [(Zheng et al, 2021)](https://arxiv.org/pdf/2104.08671.pdf).

PaperRepoDatasetLeaderboardHomepage

How to run

ltb · case-hold
$ npm i -g lettokenburn-cli
$ ltb bench pull case-hold --n 50
$ ltb bench run ./case-hold -a claude-code -m anthropic/claude-opus-4-8
$ ltb bench push

Runs happen on your machine with your own subscription; only the score summary is pushed. CLI docs

Scores

0 sourced entries

No sourced scores recorded yet.

Every number links to where it was published. Low-confidence rows are muted and need per-cell verification.

Reproduced by the community

All runs
Loading community runs…

About

CaseHOLD QA CaseHOLD is a multiple choice question answering task derived from legal citations in judicial rulings. CaseHOLD consists of ~53,000 questions, mined from the Harvard Law Library case law corpus.

Dataset repository https://huggingface.co/datasets/casehold/casehold Publication "When Does Pretraining Help? Assessing Self-Supervised Learning for Law and the CaseHOLD Dataset" ICAIL, 2021 https://reglab.stanford.edu/data/casehold-benchmark/ https://arxiv.org/abs/2104.08671

Data content The citing context from the judicial decision serves as the prompt for the question. The answer choices are holding statements derived from citations following text in a legal decision. There are five answer choices for each citing text. The correct answer is the holding statement that corresponds to the citing text. The four incorrect answers are other holding statements.