AI Paper Atlas

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

Learning Vision-and-Language Navigation from YouTube Videos

Kunyang Lin, Peihao Chen, Diwei Huang, Thomas H. Li, Mingkui Tan, Chuang Gan

ICCV 2023 · Vision-language and multimodal › Robot action and navigation · approach: Pretraining and data augmentation for VLN · 40 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
use entropy-based frame selection instead of random sampling (ablation)YouTube-VLN pre-training · R2R val unseenSR (%): 64.41 -> 66.70Improved
use entropy-based frame selection instead of temporal difference (ablation)YouTube-VLN pre-training · R2R val unseenSR (%): 65.30 -> 66.70Improved
use YouTube video frames instead of Airbnb images as data source (ablation)VLN pre-training · R2R val unseenSR (%): +0.89% SRImproved
use reasonable chronological trajectories instead of shuffled trajectories (ablation)VLN pre-training · R2R val unseenSR (%): +1.27% SRImproved
fill instructions with pseudo-labeled actions instead of random action words (ablation)action-aware instruction generator · R2R val unseenSR (%): +1.10% SR (val unseen), +0.30% (val seen)Improved
add trajectory judgment pretext task (ablation)Lily pre-training · R2R val seen / val unseenSR (%): +2.05% val seen, +0.60% val unseenImproved
pre-train DUET with YouTube-VLN and trajectory judgmentDUET · REVERIE val unseenSR (%): 46.98 -> 48.11Improved
pre-train DUET with YouTube-VLN and trajectory judgmentDUET · R2R val unseen (generative)SR (%): 72 -> 74Improved
pre-train on YouTube-VLN then fine-tune on one environmentLily · R2R val unseenSR (%): 22.43 -> 50.86 (vs VLN-BERT)Improved
Lily versus Airbert on R2R val unseenLily · R2R val unseen (discriminative)SR (%): 68.67 -> 70.00Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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