# Drive-KD: Multi-Teacher Distillation for VLMs in Autonomous Driving

arXiv: https://arxiv.org/abs/2601.21288 · HTML: https://mlsanity.com/paper/2601.21288/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Reasoning VLA and VLM driving policies, one of 9 directions in End-to-end autonomous driving.
- Builds on 47 of its 62 references that are in the map.
- First posted on arXiv 29 Jan 2026.

## Placed in

- End-to-end autonomous driving / [Reasoning VLA and VLM driving policies](https://mlsanity.com/driving/reasoning-vla-and-vlm-driving-policies/)

## Builds on (most cited first, 30 of 47)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [nuScenes: A multimodal dataset for autonomous driving](https://mlsanity.com/paper/1903.11027/) (2019)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Qwen Technical Report](https://mlsanity.com/paper/2309.16609/) (2023)
- [A Survey of Large Language Models](https://mlsanity.com/paper/2303.18223/) (2023)
- [InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks](https://mlsanity.com/paper/2312.14238/) (2023)
- [Knowledge Distillation: A Survey](https://mlsanity.com/paper/2006.05525/) (2020)
- [BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning](https://mlsanity.com/paper/1805.04687/) (2018)
- [Gradient Surgery for Multi-Task Learning](https://mlsanity.com/paper/2001.06782/) (2020)
- [Planning-oriented Autonomous Driving](https://mlsanity.com/paper/2212.10156/) (2022)
- [End to End Learning for Self-Driving Cars](https://mlsanity.com/paper/1604.07316/) (2016)
- [Vision-Language Models for Vision Tasks: A Survey](https://mlsanity.com/paper/2304.00685/) (2023)
- [DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models](https://mlsanity.com/paper/2512.02556/) (2025)
- [HallusionBench: An Advanced Diagnostic Suite for Entangled Language Hallucination and Visual Illusion in Large Vision-Language Models](https://mlsanity.com/paper/2310.14566/) (2023)
- [End-to-end Autonomous Driving: Challenges and Frontiers](https://mlsanity.com/paper/2306.16927/) (2023)
- [DriveVLM: The Convergence of Autonomous Driving and Large Vision-Language Models](https://mlsanity.com/paper/2402.12289/) (2024)
- [DriveGPT4: Interpretable End-to-end Autonomous Driving via Large Language Model](https://mlsanity.com/paper/2310.01412/) (2023)
- [A Survey of Autonomous Driving: Common Practices and Emerging Technologies](https://mlsanity.com/paper/1906.05113/) (2019)
- [LMDrive: Closed-Loop End-to-End Driving with Large Language Models](https://mlsanity.com/paper/2312.07488/) (2023)
- [A Survey on Knowledge Distillation of Large Language Models](https://mlsanity.com/paper/2402.13116/) (2024)
- [A Survey of Motion Planning and Control Techniques for Self-driving Urban Vehicles](https://mlsanity.com/paper/1604.07446/) (2016)
- [Self-Driving Cars: A Survey](https://mlsanity.com/paper/1901.04407/) (2019)
- [LLaVA-Phi: Efficient Multi-Modal Assistant with Small Language Model](https://mlsanity.com/paper/2401.02330/) (2024)
- [Vision Language Models in Autonomous Driving: A Survey and Outlook](https://mlsanity.com/paper/2310.14414/) (2023)
- [A Survey of End-to-End Driving: Architectures and Training Methods](https://mlsanity.com/paper/2003.06404/) (2020)
- [Towards Interpreting Visual Information Processing in Vision-Language Models](https://mlsanity.com/paper/2410.07149/) (2024)
- [Are VLMs Ready for Autonomous Driving? An Empirical Study from the Reliability, Data, and Metric Perspectives](https://mlsanity.com/paper/2501.04003/) (2025)
- [Less is More: Task-aware Layer-wise Distillation for Language Model Compression](https://mlsanity.com/paper/2210.01351/) (2022)

## Abstract (arXiv)

Autonomous driving is an important and safety-critical task, and recent advances in LLMs/VLMs have opened new possibilities for reasoning and planning in this domain. However, large models demand substantial GPU memory and exhibit high inference latency, while conventional supervised fine-tuning (SFT) often struggles to bridge the capability gaps of small models. To address these limitations, we propose Drive-KD, a framework that decomposes autonomous driving into a "perception-reasoning-planning" triad and transfers these capabilities via knowledge distillation. We identify layer-specific attention as the distillation signal to construct capability-specific single-teacher models that outperform baselines. Moreover, we unify these single-teacher settings into a multi-teacher distillation framework and introduce asymmetric gradient projection to mitigate cross-capability gradient conflicts. Extensive evaluations validate the generalization of our method across diverse model families and scales. Experiments show that our distilled InternVL3-1B model, with ~42 times less GPU memory and ~11.4 times higher throughput, achieves better overall performance than the pretrained 78B model from the same family on DriveBench, and surpasses GPT-5.1 on the planning dimension, providing insights toward efficient autonomous driving VLMs.