US 12,613,874 B2 · Executing queries with hallucination safeguards

Abstract

The described system aims to reduce or eliminate inaccuracies and hallucinations in responses generated by a machine learning model when processing user queries. The data platform parses and categorizes the text within data files to create structured textual representations. The user submits multiple prompts which are collectively assessed to refine and modify the initial queries. The modified query is used to identifying segments of data files that are most relevant to the query. These relevant portions are then compiled into a Retrieval-Augmented Generation (RAG) context block. This RAG context block is fed into a prompt response machine learning model, which processes the enriched information to generate a well-informed and accurate response to the user's query. Finally, this response is displayed back to the user through the chat interface, completing a cycle that enhances the reliability and relevance of machine-generated answers.

Claims20

Dependent claims

A system comprising:

at least one hardware processor;
and
at least one memory storing instructions that cause the at least one hardware processor to perform operations comprising:
generating textual representations of data files by parsing and categorizing text of the data files;
causing display of a chat message within a user interface configured to receive prompts from a first user;
receiving a plurality of prompts from the first user via the user interface, the plurality of prompts comprising a first query;
generating a modified first query based on the plurality of prompts;
identifying relevant portions of the textual representations based on the modified first query;
generating a retrieval-augmented generation (RAG) context block based on the relevant portions of the textual representations;
inputting the RAG context block into a prompt response machine learning model to generate a response to the first query, the prompt response machine learning model being trained to generate responses to queries based on inputted RAG context blocks;
and
causing display of the response to the first query to the first user within the user interface.

9 more elements

Dependencies are inferred from the claim text.

Executing queries with hallucination safeguards

Life of this patent

Executing queries with hallucination safeguards

Life of this patent

Abstract

Claims20