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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

A longstanding challenge for self-driving development is simulating dynamicdriving scenarios seeded from recorded driving logs. In pursuit of thisfunctionality, we apply tools from discrete sequence modeling to model howvehicles, pedestrians and cyclists interact in driving scenarios. Using asimple data-driven tokenization scheme, we discretize trajectories tocentimeter-level resolution using a small vocabulary. We then model themulti-agent sequence of motion tokens with a GPT-like encoder-decoder that isautoregressive in time and takes into account intra-timestep interactionbetween agents. Scenarios sampled from our model exhibit state-of-the-artrealism; our model tops the Waymo Sim Agents Benchmark, surpassing prior workalong the realism meta metric by 3.3% and along the interaction metric by 9.9%.We ablate our modeling choices in full autonomy and partial autonomy settings,and show that the representations learned by our model can quickly be adaptedto improve performance on nuScenes. We additionally evaluate the scalability ofour model with respect to parameter count and dataset size, and use densityestimates from our model to quantify the saliency of context length andintra-timestep interaction for the traffic modeling task.

Trajeglish: Learning the Language of Driving Scenarios