BROWSE
Detail
Conference Paper
GPT 모델과 Chain-of-thought 프롬프팅을 활용한 SayCan tabletop 환경에서 로봇작업 계획수립 성능 분석
Cited - time in 
Share 
Share
- Issue Date
- 2023-06
- Citation
- 대한전자공학회 학술 대회 (하계) 2023, pp.1320-1324
- Publisher
- 대한전자공학회
- Language
- Korean
- Type
- Conference Paper
- Abstract
- In this paper, we investigate the impact of Chainof-thought prompting on the SayCan tabletop manipulation task using a large language model. We show that chain-of-thought prompting leads to a 6.25% performance improvement over standard prompting using GPT-3 text-ada-001 model in the SayCan tabletop environment. These results demonstrate the potential of prompting strategies to optimize the performance of robotic task planning in a variety of scenarios in the SayCan tabletop environment.
- KSP Keywords
- Language model, Task planning, performance improvement