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

For robots to be generally useful, they must be able to find arbitraryobjects described by people (i.e., be language-driven) even without expensivenavigation training on in-domain data (i.e., perform zero-shot inference). Weexplore these capabilities in a unified setting: language-driven zero-shotobject navigation (L-ZSON). Inspired by the recent success of open-vocabularymodels for image classification, we investigate a straightforward framework,CLIP on Wheels (CoW), to adapt open-vocabulary models to this task withoutfine-tuning. To better evaluate L-ZSON, we introduce the Pasture benchmark,which considers finding uncommon objects, objects described by spatial andappearance attributes, and hidden objects described relative to visibleobjects. We conduct an in-depth empirical study by directly deploying 21 CoWbaselines across Habitat, RoboTHOR, and Pasture. In total, we evaluate over 90knavigation episodes and find that (1) CoW baselines often struggle to leveragelanguage descriptions, but are proficient at finding uncommon objects. (2) Asimple CoW, with CLIP-based object localization and classical exploration --and no additional training -- matches the navigation efficiency of astate-of-the-art ZSON method trained for 500M steps on Habitat MP3D data. Thissame CoW provides a 15.6 percentage point improvement in success over astate-of-the-art RoboTHOR ZSON model.

CoWs on Pasture: Baselines and Benchmarks for Language-Driven Zero-Shot Object Navigation