English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Visual language navigation (VLN) is an embodied task demanding a wide rangeof skills encompassing understanding, perception, and planning. For such amultifaceted challenge, previous VLN methods totally rely on one model's ownthinking to make predictions within one round. However, existing models, eventhe most advanced large language model GPT4, still struggle with dealing withmultiple tasks by single-round self-thinking. In this work, drawing inspirationfrom the expert consultation meeting, we introduce a novel zero-shot VLNframework. Within this framework, large models possessing distinct abilitiesare served as domain experts. Our proposed navigation agent, namely DiscussNav,can actively discuss with these experts to collect essential information beforemoving at every step. These discussions cover critical navigation subtasks likeinstruction understanding, environment perception, and completion estimation.Through comprehensive experiments, we demonstrate that discussions with domainexperts can effectively facilitate navigation by perceivinginstruction-relevant information, correcting inadvertent errors, and siftingthrough in-consistent movement decisions. The performances on therepresentative VLN task R2R show that our method surpasses the leadingzero-shot VLN model by a large margin on all metrics. Additionally, real-robotexperiments display the obvious advantages of our method over single-roundself-thinking.

Discuss Before Moving: Visual Language Navigation via Multi-expert Discussions