Dolphins: Multimodal Language Model for Driving

TL;DR

Dolphins is a multimodal vision-language model designed for autonomous driving, capable of understanding complex scenarios and performing diverse tasks with human-like reasoning and adaptability.

Contribution

The paper introduces Dolphins, a novel multimodal driving assistant model that incorporates a Grounded Chain of Thought process and domain-specific instruction tuning.

Findings

Enhanced reasoning with Grounded Chain of Thought

Holistic understanding of complex driving scenarios

Emergence of human-like adaptive capabilities

Abstract

The quest for fully autonomous vehicles (AVs) capable of navigating complex real-world scenarios with human-like understanding and responsiveness. In this paper, we introduce Dolphins, a novel vision-language model architected to imbibe human-like abilities as a conversational driving assistant. Dolphins is adept at processing multimodal inputs comprising video (or image) data, text instructions, and historical control signals to generate informed outputs corresponding to the provided instructions. Building upon the open-sourced pretrained Vision-Language Model, OpenFlamingo, we first enhance Dolphins's reasoning capabilities through an innovative Grounded Chain of Thought (GCoT) process. Then we tailored Dolphins to the driving domain by constructing driving-specific instruction data and conducting instruction tuning. Through the utilization of the BDD-X dataset, we designed and consolidated four distinct AV tasks into Dolphins to foster a holistic understanding of intricate driving scenarios. As a result, the distinctive features of Dolphins are characterized into two dimensions: (1) the ability to provide a comprehensive understanding of complex and long-tailed open-world driving scenarios and solve a spectrum of AV tasks, and (2) the emergence of human-like capabilities including gradient-free instant adaptation via in-context learning and error recovery via reflection.