CALVIN

Overview edit

Variants: CALVIN

Associated Benchmarks

This dataset is used in 2 benchmarks:

• Robot Manipulation - Metrics: avg. sequence length (D to D)
• Zero-shot Generalization - Metrics: Avg. sequence length

Recent Benchmark Submissions

Research Papers

Recent papers with results on this dataset:

• DreamVLA: A Vision-Language-Action Model Dreamed with Comprehensive World Knowledge (2025) -
• UniVLA: Learning to Act Anywhere with Task-centric Latent Actions (2025) -
• OpenHelix: A Short Survey, Empirical Analysis, and Open-Source Dual-System VLA Model for Robotic Manipulation (2025) -
• UP-VLA: A Unified Understanding and Prediction Model for Embodied Agent (2025) -
• Video Prediction Policy: A Generalist Robot Policy with Predictive Visual Representations (2024) -
• Towards Generalist Robot Policies: What Matters in Building Vision-Language-Action Models (2024) -
• Efficient Diffusion Transformer Policies with Mixture of Expert Denoisers for Multitask Learning (2024) -
• VidMan: Exploiting Implicit Dynamics from Video Diffusion Model for Effective Robot Manipulation (2024) -
• Towards Synergistic, Generalized, and Efficient Dual-System for Robotic Manipulation (2024) -
• GR-MG: Leveraging Partially Annotated Data via Multi-Modal Goal-Conditioned Policy (2024) -
• RoboUniView: Visual-Language Model with Unified View Representation for Robotic Manipulation (2024) -
• OpenVLA: An Open-Source Vision-Language-Action Model (2024) -
• From LLMs to Actions: Latent Codes as Bridges in Hierarchical Robot Control (2024) -
• 3D Diffuser Actor: Policy Diffusion with 3D Scene Representations (2024) -
• Unleashing Large-Scale Video Generative Pre-training for Visual Robot Manipulation (2023) -
• Vision-Language Foundation Models as Effective Robot Imitators (2023) -
• Learning Universal Policies via Text-Guided Video Generation (2023) -
• RT-1: Robotics Transformer for Real-World Control at Scale (2022) -