420 datasets found

Neural sequence models can generate highly fluent sentences, but recent studies have also shown that they are also prone to hallucinate additional content not supported by the...

The Machine Translation and Automated Analysis of the Sumerian Language dataset, which contains Sumerian texts in cuneiform script.

The dataset used for the study of Sumerian cuneiform, including part-of-speech tagging, named entity recognition, and machine translation.

The dataset consists of embeddings of all n-grams of a natural language, constituting a representative sample of a language fractal structure.

The dataset used in the paper is the SNLI dataset.

Linear-time minimum Bayes risk decoding with reference aggregation

Finetuned language models are zero-shot learners

The paper presents the results of the OpenAI Codex evaluation on generating Python code.

Multi-prompt decoding for conditional text generation

A dataset for evaluating the logical reasoning ability of chatgpt and gpt-4.

Definition Extraction (DE) is one of the well-known topics in Information Extraction that aims to identify terms and their corresponding definitions in unstructured texts.

The Chimera dataset is a ‘Chimera’ dataset of (Lazaridou et al., 2017). This dataset was specifically constructed to sim- ulate a nonce situation where a speaker encoun- ters a...

Multilingual extension of the TAXINLI dataset for analyzing the effects of reasoning types on cross-lingual transfer performance.

Multilingual extension of the TAXINLI dataset for analyzing the effects of reasoning types on cross-lingual transfer performance.

Multilingual extension of the TAXINLI dataset for analyzing the effects of reasoning types on cross-lingual transfer performance.

The Corpus of Linguistic Acceptability (CoLA) is a set of 10,657 English sentences labeled as grammatical or ungrammatical from published linguistics literature.

A set of 50 prompts to evaluate execution-based evaluation for NL2Bash task

The Words2Contact dataset contains verbal instructions for humanoid robots to place support contacts.

Word2Vec is a technique for learning vector representations of words in a text corpus.

SimVerb-3500 is a large-scale evaluation set of verb similarity, providing human ratings for the similarity of 3,500 verb pairs.