SR-BEVS：稀疏恢复方法提升 Fast-BEV 的鲁棒性与精度
Sparse Recovery Bird's-Eye View Model Soup for Robust Camera-Based 3D Object Detection
一篇针对 Fast-BEV 的论文，用剪枝加重训再平均的办法，把噪声天气下的鲁棒性拉高 33.8%，推理还不变慢。
论文提出 Sparse Recovery Bird's-Eye View Model Soup（SR-BEVS），针对 Fast-BEV 等 camera-based BEV 感知模型在真实环境损坏下鲁棒性不足的问题。方法基于层级权重幅度挑选剪枝候选，随机剪枝生成多个稀疏模型，各自用随机采样的学习率重训后平均成一个稠密模型，推理开销为零。在 nuScenes 数据集上，SR-BEVS 相对 Fast-BEV 基线将损坏鲁棒性最多提升 33.8%，干净精度提升 1.8%，在传感器噪声、天气扰动和光照变化下均有改善。论文还通过 mode connectivity 分析说明剪枝带来的多样性保留了可用于稳定权重平均的低损失盆地。
Sparse Recovery Bird's-Eye View Model Soup for Robust Camera-Based 3D Object Detection
Camera-based bird's-eye view (BEV) perception methods such as Fast-BEV have demonstrated promising results with fast inference speed and competitive accuracy. However, their robustness to real-world corruptions remains insufficiently explored, posing risks for autonomous driving deployment where environmental variations are inevitable. In this paper, we propose Sparse Recovery Bird's-Eye View Model Soup (SR-BEVS), a sparse recovery learning approach that simultaneously improves clean accuracy and corruption robustness without requiring additional data or architectural modifications. Our method identifies pruning candidates based on layer-wise weight magnitudes, applies random pruning to create diverse sparse models, and retrains each model with a randomly sampled learning rate. By averaging these diverse recovered models into a single dense model, we achieve enhanced robustness against various corruptions while introducing zero inference overhead. We further provide a mode connectivity analysis demonstrating that our pruning-based diversification preserves the connected low-loss basin necessary for stable weight averaging. Experiments on the nuScenes dataset demonstrate up to 33.8% improvement in corruption robustness and a 1.8% gain in clean accuracy over the Fast-BEV baseline, with consistent improvements across sensor noise, weather distortions, and illumination changes. Our simple yet effective approach may be applicable to other BEV models.