AI Paper Atlas

Research map / Vision-language and multimodal / Robot action and navigation

MapDream: Task-Driven Map Learning for Vision-Language Navigation

Guoxin Lian, Shuo Wang, Yucheng Wang, Yongcai Wang, Maiyue Chen, Kaihui Wang, Bo Zhang, Zhizhong Su, Deying Li, Zhaoxin Fan

ICML 2026 · Vision-language and multimodal › Robot action and navigation · approach: Pretraining and data augmentation for VLN · 0 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 Stage 1 supervised generative BEV map conditioning to the VLN policy (ablation)NVILA-2B VLN policy · R2R-CE Val-UnseenSR: 37.7 -> 42.2Improved
add Stage 2 joint reinforcement fine-tuning (GRPO) of map module and policy (ablation)MapDream after SPT · R2R-CE Val-UnseenSR: 42.2 -> 45.6Improved
add Stage 2 RFT, effect on SPL and NE (ablation)MapDream after SPT · R2R-CE Val-UnseenSPL / NE: SPL 36.5 -> 40.5; NE 7.03 -> 6.35Improved
add RFT when map is initialized with Distance channel only (ablation)MapDream · R2R-CE Val-UnseenSR: 39.4 -> 44.9Improved
add RFT when map is initialized with Landmark channel only (ablation)MapDream · R2R-CE Val-UnseenSR: 40.4 -> 44.5Improved
add RFT when map is initialized with Occupancy channel only (ablation)MapDream · R2R-CE Val-UnseenSR: 39.4 -> 43.6Improved
use all three channels instead of a single channel for map supervision (ablation)MapDream after RFT · R2R-CE Val-UnseenSR: final SR 43.6-45.6 across channel choicesNo clear effect
increase BEV map size from 96x96 (36 tokens) to 384x384 (576 tokens) (ablation)MapDream map generator · R2R-CE Val-UnseenSR: 42.2 -> 43.1No clear effect
increase BEV map size from 96x96 to 384x384 (ablation)MapDream map generator · R2R-CE Val-Unseeninference latency per step (s): 1.3 -> 12.7Hurt
use MapDream with monocular RGB instead of panoramic baselinesMapDream · R2R-CE Val-UnseenSPL: 54.4 vs BEVBert-FSTTA 51.0Improved
compare MapDream against BEVBert-FEEDTTA on success rateMapDream · R2R-CE Val-UnseenSR: 59.8 vs 61.0No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 4 of 4 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