AI Paper Atlas

Research map / Label-efficient and robust learning / Knowledge distillation

AMap: Distilling Future Priors for Ahead-Aware Online HD Map Construction

Ruikai Li, Xinrun Li, Mengwei Xie, Hao Shan, Shoumeng Qiu, Xinyuan Chang, Yizhe Fan, Feng Xiong, Han Jiang, Yilong Ren, Haiyang Yu, Mu Xu and 3 more

CVPR 2026 · Label-efficient and robust learning › Knowledge distillation · approach: Patient consistent distillation · 5 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
mask forward map region fully before trajectory prediction (ablation)MapTR maps with HiVT predictor · downstream trajectory predictionminFDE: 0.7919 -> 0.9024 (+13.95%)Hurt
mask backward map region before trajectory prediction (ablation)MapTR maps with HiVT predictor · downstream trajectory predictionMiss Rate: 0.0854 -> 0.0807 at 70% maskNo clear effect
distill from a future-frame teacher into current-frame student (AMap)MapTracker R18 student · nuScenes valA-mAP: 64.63 -> 66.28Improved
add AMap distillation to MapTR R18MapTR R18 · nuScenes valA-mAP: 36.34 -> 51.29Improved
add AMap distillation on Argoverse 2MapTracker R18 student · Argoverse 2 valA-mAP: 65.29 -> 67.81Improved
use BEVDistill or MapDistill instead of AMap for future-to-static distillationMapTracker R18 student · nuScenes valmAP: 62.81 -> 57.97 (BEVDistill), 57.82 (MapDistill)Hurt
distill both basic and refined BEV features (ablation)AMap student · nuScenes valmAP: 62.81 -> 63.47Improved
distill only basic BEV features (ablation)AMap student · nuScenes valmAP: 62.81 -> 62.66No clear effect
add query-level to BEV-level distillation (ablation)AMap student · nuScenes valmAP: 63.47 -> 64.49Improved
refined-BEV-only distillation (ablation)AMap student · nuScenes valmAP vs A-mAP/R-mAP: mAP 62.81 -> 62.72, A-mAP 64.63 -> 64.68, R-mAP 64.04 -> 64.78Mixed
use Hungarian matching on query embeddings (MSE or cosine) instead of logits KL (ablation)AMap query distillation · nuScenes valmAP: 63.63 -> 62.99 (MSE), 62.43 (cosine)Hurt
use MapDistill dummy or Top-K query distillation (ablation)AMap query distillation · nuScenes valmAP: 63.63 -> 57.63 (dummy), 56.61 (Top K)Hurt

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