Develop LLM for Confidence Scoring

[arng4108]

💡 Summary

Research and develop a large language model (LLM) to be used for confidence scoring and other PE tasks.

Motivation and context

LLMs allow for the automatic analysis of input text. Some examples of insights that can be extracted from input text using an LLM include: sentiment, context, named-entity recognition, etc. These insights can then be used for various purposes including confidence scoring.

Implementation notes

• Could be used for confidence scoring
• Could be used for PE report findings
• Could be used for quality control of PE's deliverables

Acceptance criteria

How do we know when this work is done?

• Research LLMs
• Formulate a plan
• Implement LLM and train it on task specific data
• Apply trained LLM to new data and gauge accuracy
• Incorporate LLM results into confidence scoring and other tasks