CALVIN

Composing Actions from Language and Vision

Introduced in CALVIN: A Benchmark for Language-Conditioned Policy Learning for Long-Horizon Robot Manipulation Tasks2021

About this Dataset

CALVIN (Composing Actions from Language and Vision), is an open-source simulated benchmark to learn long-horizon language-conditioned robot manipulation tasks.

Source: CALVIN: A Benchmark for Language-Conditioned Policy Learning for Long-Horizon Robot Manipulation Tasks

Dataset Variants

CALVIN

Papers1

CALVIN: A Benchmark for Language-Conditioned Policy Learning for Long-Horizon Robot Manipulation Tasks

CALVIN (Composing Actions from Language and Vision), an open-source simulated benchmark to learn long-horizon language-conditioned tasks, is presented, suggesting that there is significant room for developing innovative agents that learn to relate human language to their world models with this benchmark.

Dataset Loaders

EDIT

📦

mees/calvin

Tasks

EDIT

Robot Manipulation Success Rate (5 task-horizon)Avg. sequence length Zero-shot Generalization

Similar Datasets