# Compressing Multi-Task Model for Autonomous Driving via Pruning and Knowledge Distillation

arXiv: https://arxiv.org/abs/2511.05557 · HTML: https://mlsanity.com/paper/2511.05557/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Detection and dense prediction distillation, one of 12 directions in Knowledge distillation.
- Builds on 7 of its 14 references that are in the map.
- First posted on arXiv 3 Nov 2025.

## Placed in

- Knowledge distillation / [Detection and dense prediction distillation](https://mlsanity.com/distill/detection-and-dense-prediction-distillation/)

## Builds on (most cited first, 7 of 7)

- [BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning](https://mlsanity.com/paper/1805.04687/) (2018)
- [Importance Estimation for Neural Network Pruning](https://mlsanity.com/paper/1906.10771/) (2019)
- [YOLOP: You Only Look Once for Panoptic Driving Perception](https://mlsanity.com/paper/2108.11250/) (2021)
- [Methods for Pruning Deep Neural Networks](https://mlsanity.com/paper/2011.00241/) (2020)
- [HybridNets: End-to-End Perception Network](https://mlsanity.com/paper/2203.09035/) (2022)
- [You Only Look at Once for Real-time and Generic Multi-Task](https://mlsanity.com/paper/2310.01641/) (2023)
- [RMT-PPAD: Real-time Multi-task Learning for Panoptic Perception in Autonomous Driving](https://mlsanity.com/paper/2508.06529/) (2025)

## Built on by (most cited first, 1 of 1)

- [A Closed-Loop Evaluation of Capability Loss and Recovery in Compressed Driving Policies](https://mlsanity.com/paper/2609.00718/) (2026)

## Abstract (arXiv)

Autonomous driving systems rely on panoptic perception to jointly handle object detection, drivable area segmentation, and lane line segmentation. Although multi-task learning is an effective way to integrate these tasks, its increasing model parameters and complexity make deployment on on-board devices difficult. To address this challenge, we propose a multi-task model compression framework that combines task-aware safe pruning with feature-level knowledge distillation. Our safe pruning strategy integrates Taylor-based channel importance with gradient conflict penalty to keep important channels while removing redundant and conflicting channels. To mitigate performance degradation after pruning, we further design a task head-agnostic distillation method that transfers intermediate backbone and encoder features from a teacher to a student model as guidance. Experiments on the BDD100K dataset demonstrate that our compressed model achieves a 32.7% reduction in parameters while segmentation performance shows negligible accuracy loss and only a minor decrease in detection (-1.2% for Recall and -1.8% for mAP50) compared to the teacher. The compressed model still runs at 32.7 FPS in real-time. These results show that combining pruning and knowledge distillation provides an effective compression solution for multi-task panoptic perception.