536 datasets found

A dataset for contextualised questions that requires reasoning over a context to answer.

A reading comprehension challenge that focuses on understanding causes and effects in an expository paragraph, requiring systems to apply this understanding to novel situations.

Semantic textual similarity (STS) is a fundamental NLP task that measures the semantic similarity between a pair of sentences. In order to reduce the inherent ambiguity posed...

A novel design of ontology based adaptive hidden web crawler has been proposed that downloads the Hidden web data by using ontology driven approach.

Multiple-choice reading comprehension (MCRC) is the task of selecting the correct answer from multiple options given a question and an article. The authors propose an option...

The dataset used in the paper is a set of problems for testing the safety of artificial general intelligence (AGI) systems.

The dataset used in the paper is a synthetic arithmetic questions dataset, constructed following Liu & Low (2023). The input numbers are randomly generated, ensuring a very...

This survey concentrates on few-shots In-Context Learning (ICL) using retrieved examples for large language models, a key aspect of Retrieval-Augmented Generation (RAG).

This survey concentrates on few-shots In-Context Learning (ICL) using retrieved examples for large language models, a key aspect of Retrieval-Augmented Generation (RAG).

This survey concentrates on few-shots In-Context Learning (ICL) using retrieved examples for large language models, a key aspect of Retrieval-Augmented Generation (RAG).

The RuBQ 2.0 dataset is a large-scale, complex, and natural dataset for end-to-end question answering, composed of 10,000 question-answer pairs.

The dataset used for training and testing the MST5 model, which consists of natural language questions paired with their corresponding Wikidata SPARQL queries.

The Mintaka dataset is a large-scale, complex, and natural dataset for end-to-end question answering, composed of 20,000 question-answer pairs.

A dataset for information-seeking questions that require retrieving the necessary information missing from the original context.

A multi-document QA system that uses a three-step pipeline to perform the task: decompose, retrieve, and aggregate.

A dataset for video question answering, including video evidence selection.

Entity linking (EL), the task of identifying entities and mapping them to the correct entries in a database, is crucial for analyzing factoid questions and for building robust...

Knowledge graphs (KGs) such as Freebase (Bollacker et al. 2008), DBpedia (Auer et al. 2007), and YAGO (Mahdisoltani, Biega, and Suchanek 2014) play a critical role in various...

The DocumentQA dataset is a benchmark for question answering research. It consists of questions answerable using TF-IDF for paragraph selection.

The dataset used in this paper is a collection of simple and complex SQL queries, as well as mathematical questions. The dataset is used to evaluate the task planning and tool...