Table 4 Instruction for summarizing the key-insights from each fine-tuned LLM.

From: A scientific-article key-insight extraction system based on multi-actor of fine-tuned open-source large language models

Please combine the three parts of input sentences given below to refine and create a new, more comprehensive one part of sentences.

The expected JSON structure for your response is as follows:

{

“Summarized sentences”: “[Insert analysis here]”,

}

Input Part 1: [Sentences generated by the Mixtral FT]

Input Part 2: [Sentences generated by the Yi FT]

Input Part 3: [Sentences generated by the InternLM2 FT]

Back to article page