126 results about "Continuation" patented technology

There is provided a computer-implemented method for ensuring that a large language model (LLM) generates original text, including (i) providing or accessing a database of previous text that the LLM should not generate, wherein the database includes text used to train the LLM; (ii) checking potential continuations generated by the LLM against the database; (iii) when a potential continuation generated by the LLM matches text in the database, adjusting the potential continuation generated by the LLM to no longer match that text in the database, to produce an adjusted potential continuation, and (iv) storing the adjusted potential continuation.

There is provided a computer-implemented method for ensuring that a large language model (LLM) generates original text, including (i) providing or accessing a database of previous text that the LLM should not generate, wherein the database includes text used to train the LLM; (ii) checking potential continuations generated by the LLM against the database; (iii) when a potential continuation generated by the LLM matches text in the database, adjusting the potential continuation generated by the LLM to no longer match that text in the database, to produce an adjusted potential continuation, and (iv) storing the adjusted potential continuation.

There is provided a computer-implemented method for ensuring that a large language model (LLM) generates original text, including (i) providing or accessing a database of previous text that the LLM should not generate, wherein the database includes text used to train the LLM; (ii) checking potential continuations generated by the LLM against the database; (iii) when a potential continuation generated by the LLM matches text in the database, adjusting the potential continuation generated by the LLM to no longer match that text in the database, to produce an adjusted potential continuation, and (iv) storing the adjusted potential continuation.

There is provided a computer-implemented method for ensuring that a large language model (LLM) generates original text, including (i) providing or accessing a database of previous text that the LLM should not generate, wherein the database includes text used to train the LLM; (ii) checking potential continuations generated by the LLM against the database; (iii) when a potential continuation generated by the LLM matches text in the database, adjusting the potential continuation generated by the LLM to no longer match that text in the database, to produce an adjusted potential continuation, and (iv) storing the adjusted potential continuation.

An example of an information processing system, in a race game, causes a player object to run in a first running state if there is a first operation input at a time of landing of the player object on a track path and if there is a predetermined turn operation input at a predetermined timing at or before the time of landing. The player object runs in a second running state different from the first running state if there is the first operation input at the time of landing of the player object on the track path and if there is not the turn operation input at the predetermined timing. The player object temporarily runs in a third running state that is advantageous in the race game based on a parameter that increases in accordance with continuation of the second running state.

There is provided a computer-implemented method for ensuring that a large language model (LLM) generates original text, including (i) providing or accessing a database of previous text that the LLM should not generate, wherein the database includes text used to train the LLM; (ii) checking potential continuations generated by the LLM against the database; (iii) when a potential continuation generated by the LLM matches text in the database, adjusting the potential continuation generated by the LLM to no longer match that text in the database, to produce an adjusted potential continuation, and (iv) storing the adjusted potential continuation.

There is provided a computer-implemented method for ensuring that a large language model (LLM) generates original text, including (i) providing or accessing a database of previous text that the LLM should not generate, wherein the database includes text used to train the LLM; (ii) checking potential continuations generated by the LLM against the database; (iii) when a potential continuation generated by the LLM matches text in the database, adjusting the potential continuation generated by the LLM to no longer match that text in the database, to produce an adjusted potential continuation, and (iv) storing the adjusted potential continuation.

There is provided a computer-implemented method for ensuring that a large language model (LLM) generates original text, including (i) providing or accessing a database of previous text that the LLM should not generate, wherein the database includes text used to train the LLM; (ii) checking potential continuations generated by the LLM against the database; (iii) when a potential continuation generated by the LLM matches text in the database, adjusting the potential continuation generated by the LLM to no longer match that text in the database, to produce an adjusted potential continuation, and (iv) storing the adjusted potential continuation.

The invention provides a termination strategy optimization method considering a staged task system, and relates to the field of task termination strategies. Firstly, for each stage of a staged task, it is defined that the degradation state of a system obeys a Wiener process, and the number of decision points of each stage is obtained; secondly, converting the degradation state of each stage into a discrete state model to solve the state transition probability of the system; secondly, solving the task termination cost and the task continuation cost of each decision point in different system states, and selecting a smaller value as a corresponding expected cost; and finally, selecting task termination or task continuation action based on the current decision point and the expected cost in the system state by adopting a reverse dynamic programming algorithm to obtain a task termination strategy set. According to the optimization method provided by the invention, the blank that a staged task termination strategy is not considered in related technologies is filled up, and meanwhile, the optimization problem is expressed by adopting a random dynamic programming framework, so that the cost is effectively reduced compared with a fixed threshold strategy.

There is provided a computer-implemented method for ensuring that a large language model (LLM) generates original text, including (i) providing or accessing a database of previous text that the LLM should not generate, wherein the database includes text used to train the LLM; (ii) checking potential continuations generated by the LLM against the database; (iii) when a potential continuation generated by the LLM matches text in the database, adjusting the potential continuation generated by the LLM to no longer match that text in the database, to produce an adjusted potential continuation, and (iv) storing the adjusted potential continuation.

There is provided a method of improving the operation of a generative AI large language model (LLM)-based data processing system, by operating the LLM-based system in conjunction with a non-LLM data processing system; and in which (a) the LLM-based system sends a continuation as an input to the non-LLM system, and (b) the non-LLM system (i) uses symbolic representations to perform non-statistical reasoning on the input from the LLM-based system and (ii) generates a reasoned prompt or other context.

A method and system for executing chained instructions in a resilient, decentralized architecture are disclosed. The system enables complex workflows through an asynchronous continuation mechanism. An initial instruction, executed by a first entity in a leading validation cluster, can generate one or more continuation instructions directed to other entities. These continuations are executed immediately on a “fast track” path within the leading cluster, creating a high-speed, non-blocking sequence of cross-entity actions prior to network-wide consensus. The resulting ordered sequence is then proposed to follower clusters, which perform a consensus before executing a duplicate, verifiable instance of the entire instruction chain. This separation of optimistic, continuation-driven execution from deliberate, consensus-based finality facilitates a pipelined and parallel processing model. The architecture provides the foundation for resilient microservice applications requiring complex, low-latency, and asynchronous interactions across a scalable, decentralized system.

The invention relates to a computer-implemented method for generating a configuration specification for controlling a device, comprising the following steps: - Accessing a pre-trained language model, wherein the pre-trained language model is configured to generate a configuration specification depending on provided input data, - Receiving text data as input data for the pre-trained language model, wherein the text data are natural language text data, wherein the text data specifies requirements for controlling the device, wherein the text data is combined with additional instructions to generate an input prompt, - Processing the input prompt using the pre-trained language model, wherein a configuration specification is generated as a sequence of tokens, i. wherein a first token is generated to begin the sequence of tokens, ii. wherein a further token is added to the sequence of tokens, wherein a post-processing step is applied to ensure adding a further token as valid continuation of the sequence of tokens, iii. wherein the step of adding a further token to the sequence of tokens is repeated until a stop criterion is met, - Providing the configuration specification for controlling the device.

There is provided a method of improving the operation of a generative AI large language model (LLM)-based data processing system, by operating the LLM-based system in conjunction with a non-LLM data processing system; and in which (a) the LLM-based system sends a continuation as an input to the non-LLM system, and (b) the non-LLM system (i) uses symbolic representations to perform non-statistical reasoning on the input from the LLM-based system and (ii) generates a reasoned prompt or other context.

There is provided a method of improving the operation of a generative AI large language model (LLM)-based data processing system, by operating the LLM-based system in conjunction with a non-LLM data processing system; in which the LLM-based system sends a continuation as an input to the non-LLM system; and in which the non-LLM system (a) uses symbolic representations to (i) generate factual assertions and / or (ii) generate non-statistical reasoning steps, in each case by processing the input sent from the LLM-based system and (b) stores the factual assertions and / or non-statistical reasoning steps (“stored facts and reasoning data”) in a memory for long term re-use by the LLM-based system and / or the non-LLM system.

This disclosure relates to determining compliance of a process. A processor receives a directed graph representing the process, the directed graph comprising a full execution representing a path within the process. Based on the annotation of one of the nodes having multiple possible changes, the processor branches the full execution at the one of the nodes into multiple branches, each of the multiple branches representing a continuation of the full execution for one or more of the multiple possible changes. The processor then creates a first set of mathematical expressions including constraints to represent one of the continuations, an obligation required for compliance and a domain of the variable indicative of the annotations; and executes a solver with the first set as input to create a first output indicative of a solvability of the first set. The processor then determines, based on the first output, the compliance of the process.

There is provided a method of improving the operation of a generative AI large language model (LLM)-based data processing system, by operating the LLM-based system in conjunction with a non-LLM data processing system; and in which (a) the LLM-based system sends a continuation as an input to the non-LLM system, and (b) the non-LLM system (i) uses symbolic representations to perform non-statistical reasoning on the input from the LLM-based system and (ii) generates a reasoned prompt or other context.

The invention relates to a system (1), a method and a computer program for autonomously performing a medical procedure comprising at least one step. The system (1) comprises a computing unit (3), an interface (4) for communicating with at least one actuator (5) for operating a surgical robot (6), and a user interface (2) for inputting user instructions. The computing unit (3) is adapted to autonomously control the operation of the actuator (5) to perform said at least one step of the medical procedure, to receive control instructions from said user interface (2) in the form of one of a change instruction, a stop instruction and a continuation instruction, and to cause a change operation, a stop operation or a continuation operation of the actuator according to the control instructions.

The application relates to a long text extrapolation method and system based on period continuation of a large model, which is suitable for a ROPE position coding-based Transformer architecture model and comprises the following steps: 1) adaptively modifying a ROPE strategy; firstly, the ROPE coding strategy in the source code of the Transformer architecture model needs to be positioned; secondly, the ROPE coding strategy is updated by using an Extra-PE strategy; 2) fine-tuning the Transformer architecture model; and 3) performing reasoning by using the trained Transformer architecture model. The application converges the position coding distribution domain in a high-dimensional subspace to realize the long text extrapolation capability of the large model, and can guarantee that the attention distribution of the model in the training length window is not damaged.

Disclosed in the present invention is a method for sheet continuation of an electronic certificate layout, which expands sheet continuation modes and the attributes of a rule-based sheet continuation logic script without damaging traditional template design data structures, so that when designing a template, a template designer can, according to an actual situation, select a sheet continuation mode for a layout needing repetition design; for example, when the layout is paged according to the number of metadata sets, a copy sheet continuation mode or an automatic sheet continuation mode can be selected; and when the layout is paged according to the content of metadata, a rule-based sheet continuation mode can be selected and a sheet continuation rule can be formulated. Then a program automatically generates a layout design preparation file by means of computing and on the basis of electronic certificate filling data and electronic certificate template design data. Finally, a layout electronic file is generated by means of a layout generation technology.

The application provides a kind of based on streaming AI intelligent cache and breakpoint continuation method and system, the method includes: through the real-time encoding of the operation sequence of user and corresponding data to be uploaded by encoder, generates latent state sequence;The latent state sequence is input into the preset prediction model, and the prediction result is output, the prediction result includes: user next operation intention and affected data range;According to the latent state sequence, periodically generate and persist first breakpoint state snapshot;When breakpoint continuation is needed, the second breakpoint state snapshot selected by user from the first breakpoint state snapshot is loaded, and the replay operation sequence and replay data corresponding to the second breakpoint state snapshot are replayed based on the corresponding decoder of the encoder.

There is provided a method of improving the operation of a generative AI large language model (LLM)-based data processing system, by operating the LLM-based system in conjunction with a non-LLM data processing system; and in which (a) the LLM-based system sends a continuation as an input to the non-LLM system, and (b) the non-LLM system (i) uses symbolic representations to perform non-statistical reasoning on the input from the LLM-based system and (ii) generates a reasoned prompt or other context.

To provide a game machine capable of further enhancing interest in a game.SOLUTION: A game machine can execute a second performance (a road to a climax final battle) when a first performance execution condition (acquisition of a road icon to a climax final battle) is satisfied by executing a first performance (cherry blossom accumulation notice), and can execute a fourth performance (a road to a final battle) when a second performance execution condition (display of five character icons) is satisfied by executing a third performance (a pseudo continuation performance), and a specific performance of the same type can be executed in the second and fourth performances. The game machine can execute a fifth performance (a final battle ready-to-win) without executing a particular performance (a decisive final battle ready-to-win) when a third performance execution condition is satisfied (less than 10 character icons) by executing the second performance or the fourth performance, and at the same time can execute the particular performance without executing the fifth performance when the third performance execution condition is not satisfied (10 character icons).SELECTED DRAWING: Figure 229

To provide a game machine capable of enhancing an interest in a game suitably.SOLUTION: After completion of an opening / closing execution mode with a 5R low-probability result as an opportunity, a first time-shortening state occurs. After completion of consumption of game rounds of the number of times corresponding to a ceiling, a second time-shortening state occurs. After execution of game rounds corresponding to a time-shortening result, a third time-shortening state occurs. In first to third time-shortening states, entry into a second operation hole becomes easier and at the same time a short period tends to be selected as a variation display period of game rounds. In this case, a time-shortening state under execution is continued and the third time-shortening state becomes latent when game rounds corresponding to a time-shortening result are executed in the first time-shortening state or the second time-shortening state. One is subtracted from the remaining number of continuation times of the third latent time-shortening state when game rounds are executed in a situation in which the third time-shortening state is latent. Then, the third latent time-shortening state is started when the time-shortening state under execution is completed.SELECTED DRAWING: Figure 6

To increase amusement of a game.SOLUTION: Three rows of performance patterns are allowed to stop scrolling in a definite combination after execution of a shake performance in the second unit pattern game in a pseudo continuous performance 2, the third unit pattern game is started after three rows of performance patterns go through a non-definite ready-to-win state after execution of a shake performance in the second unit pattern game in a pseudo continuous performance 3, and three rows of performance patterns stop scrolling in a definite combination in the third unit pattern game. As the result, a possibility of stopping scrolling in a jackpot combination is generated, such that amusement of a game is increased even if three rows of performance patterns do not stop scrolling in the jackpot combination after execution of the shake performance with the player's eye.SELECTED DRAWING: Figure 32

To provide a game machine capable of realizing a suitable performance mode.SOLUTION: Based on satisfaction of a predetermined setting condition, a specific state in which specific control advantageous to a player is set for a predetermined, specific period is set by specific state setting means. When continuation of specific control over a specific period is determined, a specific performance capable of suggesting that specific control continues over the specific period to a player is executed by specific performance execution means at predetermined timing at least differing from the determined timing. Accordingly, it has an effect of realizing a suitable performance mode.SELECTED DRAWING: Figure 1

1. Name of the product of this design: Graphical user interface for automatic mode machining operations of CNC machine tool control panels. 2. Purpose of the product of this design: used for operating and controlling CNC machine tools, processing machinery, etc. 3. The design highlights of this appearance design product lie in the graphical user interface content on the product's screen. 4. The picture or photo that best illustrates the key points of the design: main view. 5. The CNC machine control panel used in this design is a common structure, omitting the pitch, left and right and rear views. 6.Purpose of Graphical User Interface: It is a software interface that provides different control functions for operating CNC machine tools. 7. Description of the changing states of the graphical user interface: The main view is the homepage interface of the software, which displays the interface in manual mode by default, and is used for fixed tool setting, tool change, reference setting and other functions; Interface change state Figure 1 is the page displayed by clicking the "Auto" button in the main view, and is used for file loading, breakpoint continuation, file processing and other functions; Interface change state Figure 2 is the program list interface displayed by clicking the "Folder" icon below the "Manual" button in the interface change state Figure 1, which is used to display the file list, and the program files under the local path are displayed by default; Interface change state Figure 3 is the program editing interface displayed by clicking the "F7 Edit" button on the program list interface in the interface change state Figure 2, and the content displayed is the selected program file content; Interface change state Figure 4 is the interface displayed by clicking the "F5 Teach" button on the program editing interface in the interface change state Figure 3, and is used for positioning teaching, cutting teaching, arc teaching and other functions.

A computer implemented method for processing instructions in a multiprocessing apparatus comprises obtaining a first instruction of a first process; decoding the first instruction to detect a continua

There is provided a method of improving the operation of a generative AI large language model (LLM)-based data processing system, by operating the LLM-based system in conjunction with a non-LLM data processing system; in which (a) the LLM-based system sends a continuation as an input to the non-LLM system; and in which the non-LLM system (i) uses symbolic representations to generate non-statistical reasoning steps by processing the input from the LLM-based system and (ii) provides the reasoning steps to the LLM-based system to enable the LLM-based system to include visible reasoning steps in a revised continuation or other output.

A first instance is caused to execute software code to perform a first portion of a workflow in response to receipt of a workflow request, and performance of the first portion results in submission of an operation request to an entity. A resume workflow request is received from the entity, where the resume workflow request includes a handle to a snapshot that corresponds to a state of execution of the software code and a response to the operation request to the entity. Using the handle to the snapshot and the response to the operation request, a second instance is caused to execute the software code from the first state to perform a second portion of the workflow. A workflow result is received from an instance that executes a last portion of the workflow, and the workflow is provided result in response to the workflow request.