Chalvatzaki, Georgia ; Younes, Ali ; Nandha, Daljeet ; Le, An Thai ; Ribeiro, Leonardo F. R. ; Gurevych, Iryna (2023)
Learning to reason over scene graphs: a case study of finetuning GPT-2 into a robot language model for grounded task planning.
In: Frontiers in Robotics and AI, 2023, 10
doi: 10.26083/tuprints-00024479
Artikel, Zweitveröffentlichung, Verlagsversion
Kurzbeschreibung (Abstract)
Long-horizon task planning is essential for the development of intelligent assistive and service robots. In this work, we investigate the applicability of a smaller class of large language models (LLMs), specifically GPT-2, in robotic task planning by learning to decompose tasks into subgoal specifications for a planner to execute sequentially. Our method grounds the input of the LLM on the domain that is represented as a scene graph, enabling it to translate human requests into executable robot plans, thereby learning to reason over long-horizon tasks, as encountered in the ALFRED benchmark. We compare our approach with classical planning and baseline methods to examine the applicability and generalizability of LLM-based planners. Our findings suggest that the knowledge stored in an LLM can be effectively grounded to perform long-horizon task planning, demonstrating the promising potential for the future application of neuro-symbolic planning methods in robotics.
| Typ des Eintrags: | Artikel |
|---|---|
| Erschienen: | 2023 |
| Autor(en): | Chalvatzaki, Georgia ; Younes, Ali ; Nandha, Daljeet ; Le, An Thai ; Ribeiro, Leonardo F. R. ; Gurevych, Iryna |
| Art des Eintrags: | Zweitveröffentlichung |
| Titel: | Learning to reason over scene graphs: a case study of finetuning GPT-2 into a robot language model for grounded task planning |
| Sprache: | Englisch |
| Publikationsjahr: | 2023 |
| Ort: | Darmstadt |
| Publikationsdatum der Erstveröffentlichung: | 2023 |
| Verlag: | Frontiers Media S.A. |
| Titel der Zeitschrift, Zeitung oder Schriftenreihe: | Frontiers in Robotics and AI |
| Jahrgang/Volume einer Zeitschrift: | 10 |
| Kollation: | 15 Seiten |
| DOI: | 10.26083/tuprints-00024479 |
| URL / URN: | https://tuprints.ulb.tu-darmstadt.de/24479 |
| Zugehörige Links: | |
| Herkunft: | Zweitveröffentlichung DeepGreen |
| Kurzbeschreibung (Abstract): | Long-horizon task planning is essential for the development of intelligent assistive and service robots. In this work, we investigate the applicability of a smaller class of large language models (LLMs), specifically GPT-2, in robotic task planning by learning to decompose tasks into subgoal specifications for a planner to execute sequentially. Our method grounds the input of the LLM on the domain that is represented as a scene graph, enabling it to translate human requests into executable robot plans, thereby learning to reason over long-horizon tasks, as encountered in the ALFRED benchmark. We compare our approach with classical planning and baseline methods to examine the applicability and generalizability of LLM-based planners. Our findings suggest that the knowledge stored in an LLM can be effectively grounded to perform long-horizon task planning, demonstrating the promising potential for the future application of neuro-symbolic planning methods in robotics. |
| Freie Schlagworte: | robot learning, task planning, grounding, language models (LMs), pretrained models, scene graphs |
| Status: | Verlagsversion |
| URN: | urn:nbn:de:tuda-tuprints-244795 |
| Sachgruppe der Dewey Dezimalklassifikatin (DDC): | 000 Allgemeines, Informatik, Informationswissenschaft > 004 Informatik |
| Fachbereich(e)/-gebiet(e): | 20 Fachbereich Informatik 20 Fachbereich Informatik > Ubiquitäre Wissensverarbeitung Zentrale Einrichtungen Zentrale Einrichtungen > hessian.AI - Hessisches Zentrum für Künstliche Intelligenz |
| Hinterlegungsdatum: | 11 Sep 2023 12:35 |
| Letzte Änderung: | 18 Sep 2023 13:37 |
| PPN: | |
| Zugehörige Links: | |
| Export: | |
| Suche nach Titel in: | TUfind oder in Google |
 |
Frage zum Eintrag |
Optionen (nur für Redakteure)
 |
Redaktionelle Details anzeigen |
Drucken
Impressum