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

Large language models (LLMs) exhibit a wide range of promising capabilities-- from step-by-step planning to commonsense reasoning -- that may provideutility for robots, but remain prone to confidently hallucinated predictions.In this work, we present KnowNo, which is a framework for measuring andaligning the uncertainty of LLM-based planners such that they know when theydon't know and ask for help when needed. KnowNo builds on the theory ofconformal prediction to provide statistical guarantees on task completion whileminimizing human help in complex multi-step planning settings. Experimentsacross a variety of simulated and real robot setups that involve tasks withdifferent modes of ambiguity (e.g., from spatial to numeric uncertainties, fromhuman preferences to Winograd schemas) show that KnowNo performs favorably overmodern baselines (which may involve ensembles or extensive prompt tuning) interms of improving efficiency and autonomy, while providing formal assurances.KnowNo can be used with LLMs out of the box without model-finetuning, andsuggests a promising lightweight approach to modeling uncertainty that cancomplement and scale with the growing capabilities of foundation models.Website: https://robot-help.github.io

Robots That Ask For Help: Uncertainty Alignment for Large Language Model Planners