RynnEC: Bringing MLLMs into Embodied World

TL;DR

RynnEC is a novel multimodal large language model that enhances embodied cognition in agents by integrating region-level video interaction, achieving state-of-the-art understanding and reasoning capabilities.

Contribution

It introduces RynnEC, a compact, region-centric video model with a new benchmark, advancing embodied cognition and general-purpose cognitive core development.

Findings

State-of-the-art performance in object property understanding

Effective object segmentation and spatial reasoning

A new egocentric video pipeline for data generation

Abstract

We introduce RynnEC, a video multimodal large language model designed for embodied cognition. Built upon a general-purpose vision-language foundation model, RynnEC incorporates a region encoder and a mask decoder, enabling flexible region-level video interaction. Despite its compact architecture, RynnEC achieves state-of-the-art performance in object property understanding, object segmentation, and spatial reasoning. Conceptually, it offers a region-centric video paradigm for the brain of embodied agents, providing fine-grained perception of the physical world and enabling more precise interactions. To mitigate the scarcity of annotated 3D datasets, we propose an egocentric video based pipeline for generating embodied cognition data. Furthermore, we introduce RynnEC-Bench, a region-centered benchmark for evaluating embodied cognitive capabilities. We anticipate that RynnEC will advance the development of general-purpose cognitive cores for embodied agents and facilitate generalization across diverse embodied tasks. The code, model checkpoints, and benchmark are available at: https://github.com/alibaba-damo-academy/RynnEC