46 datasets found

The dataset used in the paper to study the inverse frequency effect (IFE) in structural priming.

A large dataset for measuring bias in language models, including nearly 600 descriptor terms across 13 different demographic axes.

The dataset used in this paper consists of three tasks: Multi-Genre Natural Language Inference (MNLI), Quora Question Pairs (QQP), and Stanford Sentiment Treebank (SST-2).

Larger language models have higher accu- racy on average, but are they better on ev- ery single instance (datapoint)?

The dataset used in the paper to test the persuasion and susceptibility scores of language models.

The dataset used in the paper is the Anthropic Helpfulness Base eval dataset.

The dataset used in the paper is the Anthropic Helpfulness Base train dataset and the Anthropic Helpfulness eval dataset.

The dataset used in this paper is a multiple choice question set that allows for the evaluation of large language models.

The dataset is used to evaluate the performance of large language models on missing item prediction tasks.

Large language models (LLMs) can suggest missing elements from items listed in a prompt, which can be used for list completion or recommendations based on users' history....

The dataset used in the paper to test the limitations of large pretrained Language Models (LMs) on arithmetic and symbolic induction tasks.

The dataset used for the experiments in the paper, consisting of 1000 benign instruction examples.

The dataset used for the experiments in the paper, consisting of 60 harmful instructions from the AdvBench dataset.

The authors used a variety of datasets for question answering, including TriviaQA, Natural Questions, CountryQA, and Jeopardy questions.

The BIG-Bench Hard dataset is derived from the original BIG-Bench evaluation suite, focusing on tasks that pose challenges to existing language models.

The dataset used in the paper is not explicitly described, but it is mentioned that it is a preference dataset for language models.

The Pile dataset contains 3.5 million samples of diverse text for language modeling.

The CAP dataset is a corpus of adjective pairs used to evaluate adjective order preferences in language models.

The dataset used in this paper is a collection of regular expressions and grammars for constraining language models.

The dataset consists of subtitles of television shows provided by the Flemish public-service broadcaster VRT. The dataset is used to train language models of spoken Dutch.