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

Long-horizon task planning is essential for the development of intelligentassistive and service robots. In this work, we investigate the applicability ofa smaller class of large language models (LLMs), specifically GPT-2, in robotictask planning by learning to decompose tasks into subgoal specifications for aplanner to execute sequentially. Our method grounds the input of the LLM on thedomain that is represented as a scene graph, enabling it to translate humanrequests into executable robot plans, thereby learning to reason overlong-horizon tasks, as encountered in the ALFRED benchmark. We compare ourapproach with classical planning and baseline methods to examine theapplicability and generalizability of LLM-based planners. Our findings suggestthat the knowledge stored in an LLM can be effectively grounded to performlong-horizon task planning, demonstrating the promising potential for thefuture application of neuro-symbolic planning methods in robotics.

Learning to Reason over Scene Graphs: A Case Study of Finetuning GPT-2 into a Robot Language Model for Grounded Task Planning