公开(公告)号：US12067366B1

公开(公告)日：2024-08-20

申请号：US18169701

申请日：2023-02-15

Applicant: Casetext, Inc.

Inventor： Jake Heller ,  Pablo Arredondo ,  Walter DeFoor ,  Ryan Walker ,  Javed Qadrud-Din

IPC: G06F40/35

CPC classification number: G06F40/35

Abstract: Text generation prompts may be determined based on an input document and a text generation prompt template. The text generation prompts may include text from the input document and questions related to the text. The text generation prompts may be sent to a remote text generation modeling system, which may respond with text generation prompt response messages including novel text portions generated by a text generation model. The text generation prompt response messages may be parsed to generate answers corresponding with the questions.

公开(公告)号：US20250053754A1

公开(公告)日：2025-02-13

申请号：US18925933

申请日：2024-10-24

Applicant: Casetext, Inc.

Inventor： Jake Heller ,  Pablo Arredondo ,  Walter DeFoor ,  Ryan Walker ,  Javed Qadrud-Din

IPC: G06F40/56 ,  G06F40/205

Abstract: A first set of text generation prompts may be determined based on an input document and a first text generation prompt template. The first text generation prompts may include an instruction to identify factual assertions in the input text. The prompt may be sent to a remote text generation modeling system, which may respond by identifying factual assertions in the input text. A second text generation prompt may be determined based on the factual assertions and a second text generation prompt template. The second text generation prompt may include an instruction to respond to the factual assertions. A response to the input text may be generated based on written responses provided by the remote text generation modeling system.

公开(公告)号：US12159119B2

公开(公告)日：2024-12-03

申请号：US18169707

申请日：2023-02-15

Applicant: Casetext, Inc.

Inventor： Jake Heller ,  Pablo Arredondo ,  Walter DeFoor ,  Ryan Walker ,  Javed Qadrud-Din

IPC: G06F40/56 ,  G06F40/205

Abstract: A first set of text generation prompts may be determined based on an input document and a first text generation prompt template. The first set of text generation prompts may include an instruction to identify factual assertions in the input text. The prompts may be sent to a remote text generation modeling system, which may respond by identifying factual assertions in the input text. A second set of text generation prompts may be determined based on the factual assertions and a second text generation prompt template. The second set of text generation prompts may include an instruction to respond to the factual assertions. A response to the input text may be generated based on written responses provided by the remote text generation modeling system.

公开(公告)号：US20240289559A1

公开(公告)日：2024-08-29

申请号：US18515025

申请日：2023-11-20

Applicant: Casetext, Inc.

Inventor： Marcin Gajek ,  Shang Gao ,  Divyanshu Murli ,  Ryan Walker ,  Walter DeFoor ,  Javed Qadrud-Din

IPC: G06F40/40 ,  G06F16/332 ,  G06F40/289

CPC classification number: G06F40/40 ,  G06F16/3329 ,  G06F40/289 ,  G06F40/205

Abstract: A query request identifying a query and a plurality of text portions for determining an answer to the query may be received. Relevance scores corresponding with respective ones of the text portions may be determined based on application of one or more machine learning models to the respective text portion and the query. A subset of the text portions may be selected based on the relevance scores. A response message to the query request including an answer to the query in natural language text generated by a large language model based on the first subset of text portions may be determined.

公开(公告)号：US12229522B2

公开(公告)日：2025-02-18

申请号：US18515025

申请日：2023-11-20

Applicant: Casetext, Inc.

Inventor： Marcin Gajek ,  Shang Gao ,  Divyanshu Murli ,  Ryan Walker ,  Walter DeFoor ,  Javed Qadrud-Din

IPC: G06F40/40 ,  G06F16/332 ,  G06F40/205 ,  G06F40/289

Abstract: A query request identifying a query and a plurality of text portions for determining an answer to the query may be received. Relevance scores corresponding with respective ones of the text portions may be determined based on application of one or more machine learning models to the respective text portion and the query. A subset of the text portions may be selected based on the relevance scores. A response message to the query request including an answer to the query in natural language text generated by a large language model based on the first subset of text portions may be determined.

公开(公告)号：US20240289561A1

公开(公告)日：2024-08-29

申请号：US18640591

申请日：2024-04-19

Applicant: Casetext, Inc.

Inventor： Javed Qadrud-Din ,  Brian O'Kelly ,  Alan deLevie ,  Ethan Blake ,  Walter DeFoor ,  Ryan Walker ,  Pablo Arredondo

IPC: G06F40/40 ,  G06F40/226 ,  G06F40/289 ,  G06N3/045 ,  G06N3/0985

CPC classification number: G06F40/40 ,  G06F40/226 ,  G06F40/289 ,  G06N3/045 ,  G06N3/0985

Abstract: Relevance scores may be determined based on text included in a document. The text may be divided into a text portions, with the relevance scores being determined based on a comparison of a text portion of the plurality of text portions with a criterion specified in natural language. A subset of the plurality of text portions may be selected based on the plurality of relevance scores, with each of the subset of the plurality of text portions having a relevance score surpassing a threshold. A criteria evaluation prompt may be sent to a remote text generation modeling system via a communication interface. The criteria evaluation prompts may include an instruction to evaluate one or more of the subset of text portions against the criterion.

公开(公告)号：US11995411B1

公开(公告)日：2024-05-28

申请号：US18329039

申请日：2023-06-05

Applicant: Casetext, Inc.

Inventor： Javed Qadrud-Din ,  Brian O'Kelly ,  Alan deLevie ,  Ethan Blake ,  Walter DeFoor ,  Ryan Walker ,  Pablo Arredondo

IPC: G06F40/20 ,  G06F40/226 ,  G06F40/289 ,  G06F40/40 ,  G06N3/045 ,  G06N3/0985

CPC classification number: G06F40/40 ,  G06F40/226 ,  G06F40/289 ,  G06N3/045 ,  G06N3/0985

Abstract: Relevance scores may be determined based on text included in a document. The text may be divided into a text portions, with the relevance scores being determined based on a comparison of a text portion of the plurality of text portions with a criterion specified in natural language. A subset of the plurality of text portions may be selected based on the plurality of relevance scores, with each of the subset of the plurality of text portions having a relevance score surpassing a threshold. A criteria evaluation prompt may be sent to a remote text generation modeling system via a communication interface. The criteria evaluation prompts may include an instruction to evaluate one or more of the subset of text portions against the criterion.

公开(公告)号：US20250005300A1

公开(公告)日：2025-01-02

申请号：US18515032

申请日：2023-11-20

Applicant: Casetext, Inc.

Inventor： Brian O'Kelly ,  Javed Qadrud-Din ,  Ryan Walker ,  Walter DeFoor ,  Pablo Arredondo

IPC: G06F40/40 ,  G06F16/33 ,  G06F16/338 ,  G06F40/174 ,  G06F40/186 ,  G06F40/205 ,  G06F40/279 ,  G06F40/289

Abstract: A regular expression prompt may be determined by combining a regular expression prompt template with input text from an input document. The regular expression prompt template may include a natural language instruction to identify one or more regular expressions from the input text and one or more fillable portions designated for filling with the input text. The regular expression prompt may be sent to a large language model for evaluation, and one or more regular expressions may be identified based on a response received from the large language model. The regular expressions may be used to disaggregate the input text, and the disaggregated text portions may be used to determine a structured document based on the input document. The structured document may be used to determine a response to a query of the input document.

公开(公告)号：US20240273309A1

公开(公告)日：2024-08-15

申请号：US18169707

申请日：2023-02-15

Applicant: Casetext, Inc.

Inventor： Jake Heller ,  Pablo Arredondo ,  Walter DeFoor ,  Ryan Walker ,  Javed Qadrud-Din

IPC: G06F40/56 ,  G06F40/205

CPC classification number: G06F40/56 ,  G06F40/205

Abstract: A first set of text generation prompts may be determined based on an input document and a first text generation prompt template. The first set of text generation prompts may include an instruction to identify factual assertions in the input text. The prompts may be sent to a remote text generation modeling system, which may respond by identifying factual assertions in the input text. A second set of text generation prompts may be determined based on the factual assertions and a second text generation prompt template. The second set of text generation prompts may include an instruction to respond to the factual assertions. A response to the input text may be generated based on written responses provided by the remote text generation modeling system.

公开(公告)号：US11972223B1

公开(公告)日：2024-04-30

申请号：US18362738

申请日：2023-07-31

Applicant: Casetext, Inc.

Inventor： Walter DeFoor ,  Ryan Walker ,  Javed Qadrud-Din ,  Pablo Arredondo

IPC: G06F40/40 ,  G06F40/205 ,  G06F40/289 ,  G06N20/00

CPC classification number: G06F40/40 ,  G06F40/289 ,  G06N20/00 ,  G06F40/205

Abstract: A system may determine relevance prompts based on input documents and a relevance prompt template and may transmit the plurality of relevance prompts to a large language model for completion. The system may receive response messages including chunk relevance scores. The system may select a subset of the input documents based on the chunk relevance scores. The system may determine query response prompts including text from the selected input documents the natural language query, and a second set of natural language instructions to address the natural language query. The system may determine a response to the natural language query based on answers determined in response to the query response prompts.