Abstract: An example operation may include one or more of executing a machine learning (ML) model with a plurality of attention heads on a training data input during an epoch to generate a predicted output, determining a difference between the predicted output an and actual output corresponding to the training data input based on a loss function that is configured to perform preferential attachment of neurons in the ML model, modifying parameter values of the ML model based on the difference, wherein the modifying comprises modifying at least one parameter value of the parameter values of the ML model to be set to zero to generate a sparse ML model, and executing the sparse ML model on an additional training data input during an additional epoch to generate an additional predicted output.

Abstract: An embodiment includes non-deterministic agent state transition behavior specification, monitoring, and correction. An embodiment establishes an agent, wherein the agent is configured to output text in response to input text. The embodiment defines an agent behavior specification for the agent. The embodiment inputs a text input to the agent and monitors the text output of the agent to detect an incorrect state transition, wherein the incorrect state transition comprises a state transition that deviates from the agent behavior specification. The embodiment applies a correction to the output text to create a corrected output text upon detecting the incorrect state transition. The embodiment reverts the agent to a previous state, the previous state preceding the state corresponding to the incorrect state transition detected. The embodiment inputs the corrected output text to the agent in the previous state to cause future behavior of the agent to align with the agent behavior specification.

Abstract: One or more systems, devices, computer program products and/or computer-implemented methods of use provided herein relate to self-reward guided autoregressive sampling for large language models (LLMs). The system can comprise a processor that can execute computer executable components stored in a memory, where the computer executable components can comprise at least one self-reward model. The at least one self-reward model can generate a score for a sentence generated by an LLM, where the score can be based on one or more tokens comprised in the sentence and an attribute associated with the at least one self-reward model. The at least one self-reward model can further alter a text generation process employed by the LLM to generate the sentence, such that respective sampling probabilities of respective tokens comprised in a vocabulary employed by the LLM to generate a new token can be updated by the LLM based on the score.

Abstract: A method, computer program product, and computer system. A generative large language model (LLM) receives a query of text and associated text data that includes N units of text for determining a context for the query. An embedding (Zquery) of the query and a vector Z including N embeddings is generated. The N elements of Z are distributed into C memory portions of the external memory unit. Each memory portion includes S memory locations, where C=CEILING (N/S). A process (a) ascertains a closest element in each memory portion that is closest to Zquery, and (b) distributes the ascertained closest elements into memory locations within a proper subset of the C memory portions, where (a) and (b) are recursively repeated until only a single closest element is ascertained from which a context is determined. A response to the query is generated based on the query and the context.

Abstract: An approach is provided for enhancing a generative large language model (LLM). An encoder, decoder, and generative associative memory network set are jointly trained to learn to store sentence encodings in a memory matrix used during decoding. The encoder and decoder are included in the generative LLM. The generative associative memory network set is included in an external memory unit. The external memory unit is external to the encoder and decoder. The generative LLM is augmented with the external memory unit in a framework that enhances the generative LLM. The external memory unit is updated with new information. Using the new information in the updated external memory unit, a knowledge update of the decoder is performed during an inference without fine-tuning or re-training the generative LLM. A response to a prompt during the inference is generated. The response is based on the knowledge update of the decoder.

Abstract: This disclosure relates to utilizing a statistical model trained on character dimensions to determine a likelihood of a person purchasing a product. The method may include obtaining user-input data of a first person (e.g., textual-input data, survey-response data, offer information, or clickstream data associated with a first person). A character profile for the first person is derived using the user-input data and a psycholinguistic lexicon. A statistical model is generated based on the derived character profile of the first person. Second user-input data associated with a second person is obtained. The second user-input is applied to the statistical model to determine an output of the model (e.g., a statistical probability value that quantifies, for example, a predicted intention of the second person to purchase a particular product).

Abstract: This disclosure relates to utilizing a statistical model trained on character dimensions to determine a likelihood of a person purchasing a product. The method may include obtaining user-input data of a first person (e.g., textual-input data, survey-response data, offer information, or clickstream data associated with a first person). A character profile for the first person is derived using the user-input data and a psycholinguistic lexicon. A statistical model is generated based on the derived character profile of the first person. Second user-input data associated with a second person is obtained. The second user-input is applied to the statistical model to determine an output of the model (e.g., a statistical probability value that quantifies, for example, a predicted intention of the second person to purchase a particular product).