302 datasets found

The GTA-V dataset is a large-scale dataset of driving scenes, consisting of 56 training videos and 21 test videos.

The dataset is used for blood glucose prediction based on continuous glucose monitoring (CGM) device readings.

The dataset used for the code-switched speech recognition task, which consists of Mandarin-English code-switched corpora.

The nvBench-Rob(nlq,schema) dataset is a testing set from nvBench-Rob, containing both NLQ variants and data schema variants, specifically designed to test the robustness of...

The nvBench-Robschema dataset is a testing set from nvBench-Rob, containing only data schema variants, specifically designed to test the robustness of models against data schema...

The nvBench-Robnlq dataset is a testing set from nvBench-Rob, containing only NLQ variants, specifically designed to test the robustness of models against NLQ variants.

The nvBench dataset is a benchmark for text-to-vis models, containing natural language questions and their corresponding data visualizations.

The nvBench-Rob dataset is a comprehensive robustness evaluation dataset for text-to-vis models, containing diverse lexical and phrasal variations based on the original...

A laboratory-created dataset with ground-truth for hyperspectral unmixing evaluation

The dataset used in the paper for subject-driven text-to-image synthesis

The BREEDS dataset is used for evaluating the proposed method.

The dataset used in the paper is the iNaturalist 2018 dataset, which is a real-world large-scale imbalanced dataset.

The dataset used in the paper for Rotowire

The dataset used in the paper for code generation

The dataset used in the paper is the HumanEval dataset, which is used to evaluate the performance of language models.

The dataset contains 10 highly ambiguous name references with 1091 documents and 74 distinct real-life authors.

The dataset contains 10 highly ambiguous name references with 1091 documents and 74 distinct real-life authors.

Parameter-efficient fine-tuning (PEFT) using labeled task data can significantly improve the performance of large language models (LLMs) on the downstream task. However, there...

A large-scale dataset of 165190 brightfield images of 32 different cell lines across 93 experimental batches.

The LinkedIn dataset used for training and testing the proposed model, containing leaked passwords.