This repository contains the official dataset of [GIFARC: Synthetic Dataset for Leveraging Human-Intuitive Analogies to Elevate AI Reasoning].
The code generating this dataset can be found in GIFARC @ Anonymous Github.
The data demonstration is also visualized in a webpage.
Despite substantial progress in deep learning, state-of-the-art models still achieve accuracy… See the full description on the dataset page:
https://huggingface.co/datasets/DumDev/gif_arc.
