Model Card for RobotxLLM-Qwen7B-SFT

This model is a fine-tuned version of unsloth/Qwen2.5-7B-Instruct designed for decision-making and control adaptation in autonomous driving systems. It was developed as part of the RSS 2025 paper Enhancing Autonomous Driving Systems with On-Board Deployed Large Language Models.

Model Details

Model Description

This model builds on Qwen2.5-7B-Instruct via LoRA based SFT. It is trained on a synthetic dataset obtained from GPT-4o of driving instructions and corresponding robot interactions with it's corresponding low-level MPC controller to adhere to natural language commands.

Developed by: ETH Zurich, PBL and IIS
Model type: Language model fine-tuned for embodied decision making
Language(s): English
License: Apache 2.0 (inherits from base model)
Finetuned from model: unsloth/Qwen2.5-7B-Instruct

Model Sources

Repository: https://github.com/ForzaETH/LLMxRobot
Paper: https://arxiv.org/abs/2504.11514