# Do Better Visual Representations Always Lead to Better End-to-End Autonomous Driving?

arXiv: https://arxiv.org/abs/2610.09695 · HTML: https://mlsanity.com/paper/2610.09695/

## Key facts

- No citing papers in the map yet (data through 7 Oct 2026).
- Placed in End-to-end system architecture, one of 9 directions in End-to-end autonomous driving.
- Builds on 15 of its 64 references that are in the map.
- First posted on arXiv 7 Oct 2026.

## Placed in

- End-to-end autonomous driving / [End-to-end system architecture](https://mlsanity.com/driving/end-to-end-system-architecture/)

## Builds on (most cited first, 15 of 15)

- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [DINOv3](https://mlsanity.com/paper/2508.10104/) (2025)
- [Paying More Attention to Attention: Improving the Performance of Convolutional Neural Networks via Attention Transfer](https://mlsanity.com/paper/1612.03928/) (2016)
- [SAM 3: Segment Anything with Concepts](https://mlsanity.com/paper/2511.16719/) (2025)
- [Depth Anything 3: Recovering the Visual Space from Any Views](https://mlsanity.com/paper/2511.10647/) (2025)
- [VADv2: End-to-End Vectorized Autonomous Driving via Probabilistic Planning](https://mlsanity.com/paper/2402.13243/) (2024)
- [Hydra-MDP++: Advancing End-to-End Driving via Expert-Guided Hydra-Distillation](https://mlsanity.com/paper/2503.12820/) (2025)
- [Latent-WAM: Latent World Action Modeling for End-to-End Autonomous Driving](https://mlsanity.com/paper/2603.24581/) (2026)
- [MTGS: Multi-Traversal Gaussian Splatting](https://mlsanity.com/paper/2503.12552/) (2025)
- [Challenger: Affordable Adversarial Driving Video Generation](https://mlsanity.com/paper/2505.15880/) (2025)
- [DVGT-2: Vision-Geometry-Action Model for Autonomous Driving at Scale](https://mlsanity.com/paper/2604.00813/) (2026)
- [Unleashing the Potential of Diffusion Models for End-to-End Autonomous Driving](https://mlsanity.com/paper/2602.22801/) (2026)
- [FLARE: Learning Future-Aware Latent Representations from Vision-Language Models for Autonomous Driving](https://mlsanity.com/paper/2601.05611/) (2026)
- [DriveVGGT: Calibration-Constrained Visual Geometry Transformers for Multi-Camera Autonomous Driving](https://mlsanity.com/paper/2511.22264/) (2025)
- [Zero-Shot Cross-City Generalization in End-to-End Autonomous Driving: Self-Supervised versus Supervised Representations](https://mlsanity.com/paper/2603.11417/) (2026)

## Abstract (arXiv)

Visual foundation models (VFMs) are increasingly integrated into end-to-end autonomous driving for their powerful representations, yet it remains unclear when these representations improve driving performance. To investigate this question, we introduce ViRA, a planner-agnostic visual representation alignment framework that keeps the planner architecture and inference cost unchanged. Our study reveals three findings: (1) VFM-guided visual representations consistently improve driving performance across diverse end-to-end planners, with gains extending to zero-shot closed-loop evaluation. (2) The choice of VFM target matters for planning performance, and alignment to a different VFM can further benefit planners with pre-trained VFM encoders. (3) Auxiliary perception supervision reduces sensitivity to VFM target selection, narrowing the EPDMS spread across five targets from 2.7 to 0.5 points and potentially compensating for less effective VFM targets. Guided by these findings, we develop ViRA-Diffusion, a diffusion-based planner trained without auxiliary perception supervision, which achieves 92.3 EPDMS on NAVSIM v2 navtest, outperforming recent methods in our comparison by at least 1.9 points. The results motivate jointly considering target selection and planner supervision when integrating VFMs into end-to-end autonomous driving. The results and demo are available at https://github.com/OpenDriveLab/ViRA.