21 datasets found

The Twins dataset is a real-world dataset used for benchmarking treatment effect estimation methods.

IOHprofiler is a tool for analyzing and comparing iterative optimization heuristics. It provides statistical evaluations of the algorithms' performance by means of the...

A collection of 40 synthetic datasets designed to test the performance of common machine learning algorithms.

A tool for collecting traces from any Python codebase that uses OpenAI’s API.

A dataset of traces from the MACHIAVELLI environment, including API calls and their outcomes.

A structured collection of tests for input-output safeguards, including established failure tests, emerging failure tests, and next-gen architecture tests.

A comprehensive evaluation of proprietary and open-weights large language models using the first astronomy-specific benchmarking dataset.

The dataset used in this paper is a set of benchmark problems for gray-box optimization, including the Sphere function, Rosenbrock function, REBGrid function, and others.

A benchmark dataset for non-lambertian and uncalibrated photometric stereo.

The task of Question Answering over Linked Data (QALD) has received increased attention over the last years (see the surveys [14] and [36]). The task consists in mapping natural...

Human vision is naturally more attracted by some regions within their field of view than others. This intrinsic selectivity mechanism, so-called visual attention, is influenced...

A benchmark for large-scale global optimization, featuring composite functions with varying sizes and complexities.

A benchmark for real-parameter black-box optimization, focusing on noisy optimization problems.

A broad benchmark suite for black-box optimization, covering a wide range of problems, including academic benchmarks, real-world applications, and discrete optimization problems.

ConvBench is a comprehensive benchmark for evaluating and comparing convolution algorithms. It assesses 9243 convolution operations derived from 1097 real-world deep learning...

The YCB object and model set is a benchmark for manipulation research, consisting of 15 object categories and 3D models.

Benchmarking single-image dehazing and beyond

The proposed system is composed of three neural networks (NN): an auto-regressive module (ARM) fψ, an upsampler fυ and a synthesis fθ.

The MoleculeNet dataset is a benchmarking platform for molecular machine learning.

The proposed system is composed of three neural networks (NN): an auto-regressive module (ARM) fψ, an upsampler fυ and a synthesis fθ.