339 results
A multi-hazard, multi-sensor, and multi-task vision-language dataset for global-scale disaster assessment and response.
VectorOS Vector 100k SimSat VLM Dataset
UAVReason VQA / Caption / Generation
MapEval-Visual This dataset was introduced in MapEval: A Map-Based Evaluation of Geo-Spatial Reasoning in Foundation Models Example Query…
A multi-hazard, multi-sensor, and multi-task vision-language dataset for global-scale disaster assessment and response.
A 2k subset of the validation split of the HRVQA dataset ported to HF for ease-of-use in quick…
A 2k subset of the validation split of the RSVQA LR dataset ported to HF for ease-of-use in…
A 2k subset of the validation split of the RSVQA HR dataset ported to HF for ease-of-use in…
LUCID – Lunar Captioned Image Dataset
Blackline Atlas Training Corpus v1 Blackline Atlas Training Corpus v1 is a license-aware, normalized dataset for building structured…
BEAF: Before-After Changes for Hallucination Evaluation BEAF is a benchmark for evaluating object hallucination in vision-language models using…
MultiChartQA This repository contains the questions and answers for our Multi-chart Benchmark. At present, only the data is…
DocVQA 2026 ICDAR2026 Competition on Multimodal Reasoning over Documents in Multiple Domains Building upon previous DocVQA benchmarks, this…