AI Paper Atlas

Research map / Label-efficient and robust learning / Test-time adaptation

Improving Batch Normalization with Test-Time Adaptation for Robust Object Detection in Self-Driving

Dacheng Liao, Mengshi Qi, Liang Liu, Huadong Ma

AAAI 2026 · Label-efficient and robust learning › Test-time adaptation · approach: Test-time adaptation of vision-language models · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add GCM loss to LearnableBN with EM loss (ablation)LearnableBN (BEVFormer) · nuScenes-C SnowNDS: 0.2509 -> 0.2687Improved
add GCM loss to LearnableBN with EM loss (ablation)LearnableBN (BEVFormer) · nuScenes-C BrightnessNDS: 0.4583 -> 0.4680Improved
add semantic-consistency dual-stage adaptation (ablation)LearnableBN (BEVFormer) · nuScenes-C BrightnessNDS: 0.4680 -> 0.4835Improved
add semantic-consistency dual-stage adaptation (ablation)LearnableBN (BEVFormer) · nuScenes-C Color QuantNDS: 0.4099 -> 0.4192Improved
add semantic-consistency dual-stage adaptation (ablation)LearnableBN (BEVFormer) · nuScenes-C SnowNDS: 0.2687 -> 0.2718Improved
use entropy minimization loss only in LearnableBN (ablation)LearnableBN (BEVFormer) · nuScenes-C (six corruptions)NDS: Snow 0.2297 -> 0.2509; Brightness 0.4908 -> 0.4583Mixed
replace EMA BN statistics with learnable mixture coefficient (LearnableBN) vs ReviseBNBEVFormer R101 · nuScenes-C Low LightNDS: 0.2631 -> 0.3671Improved
apply TENT (entropy minimization on BN affine params) to BEVFormerBEVFormer R101 · nuScenes-C Low LightNDS: 0.4011 -> 0.2636 (easy)Hurt
apply AdaBn and ARM BN adaptation at batch size 1BEVFormer R101 · nuScenes-C Low LightNDS: 0.3212 -> 0.0772 (AdaBn)Hurt
apply LearnableBN to Sparse4DSparse4D R101 · nuScenes-C Motion BlurNDS: 0.3563 -> 0.4035Improved
apply LearnableBN to MonoFlex on KITTI-CMonoFlex · KITTI-C PedestrianAP3D|R40: Contrast 2.08 -> 14.74; Fog 2.25 -> 9.62Mixed
apply LearnableBN when baseline already strong (fog)BEVFormer R101 · nuScenes-C FogNDS: 0.4796 -> 0.4816No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy