Declarative Code Prompting via Imperative LLM Translation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional large language models, despite their sophistication, are expensive and primarily designed for imperative code, limiting their application to declarative code, which is often used by users who may not be experienced programmers.

Innovation Solution

Utilizing a language model trained on imperative code to perform tasks on declarative code by converting portions of declarative code and task prompts into imperative code, allowing the model to generate responses that modify or augment the declarative code.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If a user equipment transmits a buffer status report indicating available buffer space for downlink data, then the network can allocate downlink resources more efficiently, but the user equipment may waste power by monitoring for downlink data that will not be transmitted

Engineering Contradiction:

Improvedownlink resource allocation efficiencyVSAvoiduser equipment power consumption

Core Design Contradiction:

ProductivityVSUse of energy by moving object

Solution Approach 1:

The network device sends feedback information to the user equipment indicating whether downlink data will be transmitted. This feedback mechanism allows the user equipment to adjust its monitoring behavior based on actual downlink data transmission status, avoiding unnecessary power consumption while maintaining efficient resource allocation

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The user equipment autonomously adjusts its downlink data monitoring behavior based on the buffer status report and received feedback. When indicated that no downlink data will be transmitted, the equipment stops monitoring, thereby self-managing its power consumption without continuous network control

Inventive Principle:

Principle #25Self-service

2Use of energy by moving object

If the user equipment stops monitoring for downlink data to save power, then power consumption is reduced, but the equipment may miss downlink data transmissions

Engineering Contradiction:

Improveuser equipment power consumptionVSAvoiddownlink data reception reliability

Core Design Contradiction:

Use of energy by moving objectVSReliability

Solution Approach 1:

The network device provides explicit feedback information to the user equipment about whether downlink data will be transmitted. This feedback ensures that the user equipment only stops monitoring when it is safe to do so, maintaining reliable data reception while achieving power savings

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The network device determines in advance whether downlink data will be transmitted and communicates this information to the user equipment before the actual transmission decision is made. This preliminary information allows the user equipment to proactively adjust its monitoring state, preventing data loss while saving power

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentEP4544396B1Responding to task prompt on declarative code using language model

Publication Date: 2026.04.29 MICROSOFT TECHNOLOGY LICENSING LLC

Figure 1
Figure 2
Figure 3

AI summary

The generation of a response to a task prompt that represents a task to perform on declarative code. The response is generated with the aid of a language model that was trained on imperative code. The declarative code includes declarations about data. A task prompt represents a task to perform on the declarative code. At least a portion of the declarative code and at least a portion of the task prompt are converted into input imperative code. The input imperative code is then caused to be provided as input to the language model, resulting in the language model generating output imperative code. At least a portion of the output imperative code is then converted into a response to the task prompt.