# US 2025/0219970 A1 · Contextual conversational user assistance

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2025/0219970 A1
- Document ID: US20250219970A1
- Status: Published application
- Kind: A1
- Application number: 18/402,798
- Filed: 2024-01-03
- Published: 2025-07-03
- Earliest claimed: 2024-01-03
- Assignees: International Business Machines Corporation
- Applicants: International Business Machines Corporation
- Inventors: Logan Bailey; Jeremy R. Fox; Zachary A. Silverstein; Justin David Weisz
- Web page: https://patentgrep.com/patent/US20250219970A1

## CPC

- H04L 51/02: using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
- G06F 40/30: Semantic analysis
- G06F 40/35: Discourse or dialogue representation
- G06F 40/40: Processing or translation of natural language (natural language analysis G06F40/20; semantic analysis G06F40/30)
- G06Q 10/063114: {Status monitoring or status determination for a person or group}
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR

## Abstract

In an approach for providing contextual conversational user assistance, a processor monitors a plurality of inputs by a user to a computing device. Responsive to determining, based on the monitoring, that an error rate associated with the plurality of inputs has exceeded a threshold level, a processor invokes a conversational large learning model (LLM). A processor queries the user regarding a mental state and a task of the user using the conversational LLM. A processor identifies one or more methods to assist the user based on a response of the user to the querying. A processor executes the one or more methods identified to assist the user.

## Claims

### Claim 1

A computer-implemented method comprising:
monitoring, by one or more processors, a plurality of inputs by a user to a computing device;
responsive to determining, based on the monitoring, that an error rate associated with the plurality of inputs has exceeded a threshold level, invoking, by the one or more processors, a conversational large learning model (LLM);
querying, by the one or more processors, the user regarding a mental state and a task of the user using the conversational LLM;
identifying, by the one or more processors, one or more methods to assist the user based on a response of the user to the querying; and
executing, by the one or more processors, the one or more methods identified to assist the user.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, further comprising:
subsequent to invoking the conversational large learning model, detecting, by the one or more processors, a sentiment of the user; and
selecting, by the one or more processors, a persona of a plurality of personas for the conversational LLM to query the user based on the detected sentiment of the user.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, further comprising;
subsequent to querying the user regarding the mental state and the task of the user using the conversational LLM, processing, by the one or more processors, the response for an indication of frustration of the user; and
processing, by the one or more processors, the response for an indication of fatigue of the user.

### Claim 4 (depends on 3)

The computer-implemented method of claim 3, wherein the indication of frustration includes at least one of a Natural Language component, a speech, a language choice, a sentiment, and an expletive, and wherein the indication of fatigue includes at least one of a spelling error, a process indicator, and a process blind indicator.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, further comprising:
analyzing, by the one or more processors, one or more factors associated with an action the user is executing.

### Claim 6 (depends on 5)

The computer-implemented method of claim 5, wherein the one or more factors associated with the action the user is executing include at least one of a degree of criticality of the action the user is executing, a timeline associated with the action the user is executing, and an urgency of the action the user is executing.

### Claim 7 (depends on 1)

The computer-implemented method of claim 1, wherein the one or more methods identified to assist the user comprise at least one of helping the user with the task, chatting with the user to provide emotional support, and recommending that the user take a break.

### Claim 8

A computer program product comprising:
one or more computer readable storage media and program instructions stored on the one or more computer readable storage media, the program instructions comprising:
program instructions to monitor a plurality of inputs by a user to a computing device;
responsive to determining, based on the monitoring, that an error rate associated with the plurality of inputs has exceeded a threshold level, program instructions to invoke a conversational large learning model (LLM);
program instructions to query the user regarding a mental state and a task of the user using the conversational LLM;
program instructions to identify one or more methods to assist the user based on a response of the user to the querying; and
program instructions to execute the one or more methods identified to assist the user.

### Claim 9 (depends on 8)

The computer program product of claim 8, further comprising:
subsequent to invoking the conversational large learning model, program instructions to detect a sentiment of the user; and
program instructions to select a persona of a plurality of personas for the conversational LLM to query the user based on the detected sentiment of the user.

### Claim 10 (depends on 8)

The computer program product of claim 8, further comprising;
subsequent to querying the user regarding the mental state and the task of the user using the conversational LLM, program instructions to process the response for an indication of frustration of the user; and
program instructions to process the response for an indication of fatigue of the user.

### Claim 11 (depends on 10)

The computer program product of claim 10, wherein the indication of frustration includes at least one of a Natural Language component, a speech, a language choice, a sentiment, and an expletive, and wherein the indication of fatigue includes at least one of a spelling error, a process indicator, and a process blind indicator.

### Claim 12 (depends on 8)

The computer program product of claim 8, further comprising:
program instructions to analyze one or more factors associated with an action the user is executing.

### Claim 13 (depends on 12)

The computer program product of claim 12, wherein the one or more factors associated with the action the user is executing include at least one of a degree of criticality of the action the user is executing, a timeline associated with the action the user is executing, and an urgency of the action the user is executing.

### Claim 14 (depends on 8)

The computer program product of claim 8, wherein the one or more methods identified to assist the user comprise at least one of helping the user with the task, chatting with the user to provide emotional support, and recommending that the user take a break.

### Claim 15

A computer system comprising:
one or more computer processors;
one or more computer readable storage media;
program instructions collectively stored on the one or more computer readable storage media for execution by at least one of the one or more computer processors, the stored program instructions comprising:
program instructions to monitor a plurality of inputs by a user to a computing device;
responsive to determining, based on the monitoring, that an error rate associated with the plurality of inputs has exceeded a threshold level, program instructions to invoke a conversational large learning model (LLM);
program instructions to query the user regarding a mental state and a task of the user using the conversational LLM;
program instructions to identify one or more methods to assist the user based on a response of the user to the querying; and
program instructions to execute the one or more methods identified to assist the user.

### Claim 16 (depends on 15)

The computer system of claim 15, further comprising:
subsequent to invoking the conversational large learning model, program instructions to detect a sentiment of the user; and
program instructions to select a persona of a plurality of personas for the conversational LLM to query the user based on the detected sentiment of the user.

### Claim 17 (depends on 15)

The computer system of claim 15, further comprising;
subsequent to querying the user regarding the mental state and the task of the user using the conversational LLM, program instructions to process the response for an indication of frustration of the user; and
program instructions to process the response for an indication of fatigue of the user.

### Claim 18 (depends on 17)

The computer system of claim 17, wherein the indication of frustration includes at least one of a Natural Language component, a speech, a language choice, a sentiment, and an expletive, and wherein the indication of fatigue includes at least one of a spelling error, a process indicator, and a process blind indicator.

### Claim 19 (depends on 15)

The computer system of claim 15, further comprising:
program instructions to analyze one or more factors associated with an action the user is executing.

### Claim 20 (depends on 19)

The computer system of claim 19, wherein the one or more factors associated with the action the user is executing include at least one of a degree of criticality of the action the user is executing, a timeline associated with the action the user is executing, and an urgency of the action the user is executing.
