amazon-science/auto-cot
Automatic chain-of-thought prompting system that reduces manual effort in designing reasoning demonstrations for large language models.
Not currently ranked — collecting fresh signals.
star history
Auto-CoT automatically generates diverse chain-of-thought demonstrations to prompt large language models, removing the need for manual prompt design. It constructs reasoning chains step by step and selects demographically varied examples to cover different reasoning patterns. The approach achieves performance comparable to or exceeding manually designed prompts on GPT-3 across reasoning benchmarks like arithmetic and commonsense questions.
Frequently asked
- What is amazon-science/auto-cot?
- Automatic chain-of-thought prompting system that reduces manual effort in designing reasoning demonstrations for large language models.
- Is auto-cot open source?
- Yes — amazon-science/auto-cot is open source, released under the Apache-2.0 license.
- What language is auto-cot written in?
- amazon-science/auto-cot is primarily written in Jupyter Notebook.
- How popular is auto-cot?
- amazon-science/auto-cot has 2k stars on GitHub.
- Where can I find auto-cot?
- amazon-science/auto-cot is on GitHub at https://github.com/amazon-science/auto-cot.