The power behind V7 Go’s workflows is the ability to break down complex tasks into reasoning steps. Column by column, AI can reflect on parts of a problem rather than the whole, leading to a 32% reduction in errors over one-shot reasoning across document benchmarks, thanks to Chain of Thought Reasoning and Index Knowledge.
“V7 Go achieved 98% accuracy with a zero-shot approach using Index knowledge, against 66% and 42%.”
V7 Go is powered by Index Knowledge, a technology that breaks down large files containing information into small searchable indexes that enable LLMs to query information more accurately than retrieval augmented generation (RAG) techniques at the expense of more compute.
Traditional chat AI products simply inject text from documents into a prompt, whilst Index Knowledge utilizes the model itself to develop a data extraction plan, much like a human analyst would. This feature is particularly effective in files containing many numbers and uncommon terms such as proteins, proper nouns, and formulas, which RAG often fails to retrieve when summarizing.
We compared the extraction of these customs form against two leading document processing providers, and achieved 98% accuracy with a zero-shot approach using Index knowledge, against 66% and 42%.
Alberto Rizzoli
Co-founder and CEO
