The Cornell Natural Language Visual Reasoning (NLVR) is a language grounding dataset. It contains 92,244 pairs of natural language statements grounded in synthetic images. The task is to determine whether a sentence is true or false about an image. The data was collected through crowdsourcing, and requires reasoning about sets of objects, quantities, comparisons, and spatial relations.
Have questions? Please contact Alane Suhr (suhr <at> cs.cornell.edu).
|There is exactly one black triangle not touching any edge||true|
|there is at least one tower with four blocks with a yellow block at the base and a blue block below the top block||true|
|There is a box with multiple items and only one item has a different color.||false|
|There is exactly one tower with a blue block at the base and yellow block at the top||false|
The data is split into training, development, and two test sets. The first test set is public and available with the data, the second will not be released. The ranking in the leaderboards below is based on results on the unreleased test set. We will soon post instruction for submitting systems to test with the unreleased data. In the meantime, if you are interested in testing your system, please email the authors.
|Date||Model||Development||Public Test||Unreleased Test||2017.04.22||Neural Module Networks (Andreas et. al. 2016), details in Suhr et. al. 2017||63.06%||66.12%||61.99%|
|Date||Model||Development||Public Test||Unreleased Test|
|2017.04.22||MaxEnt on sent+img features, details in Suhr et. al. 2017||68.04%||67.68%||67.82%|