Abstract: Embodiments disclosed are directed to a computing system that performs steps to automatically identify risk control features and entities in a risk control document. The computing system regenerates, by a semantic prediction machine learning (ML) model, phrases in a risk control document. The computing system then classifies, by the semantic prediction ML model, risk control features associated with the regenerated phrases. Subsequently, the computing system corrects, by a discriminative natural language processing (NLP) model, the classified risk control features based on the phrases and the regenerated phrases.

Abstract: Methods, systems, and apparatuses are described herein for protecting the privacy of parties conducting Non-Fungible Token (NFT) transfers by conducting separate NFT transactions on a private blockchain separate from a public blockchain. An issuer and recipient may generate a token secret, and the issuer may send a create token transaction request comprising a unique token identifier, the token secret, and zero-knowledge proof data. Based on that request, an NFT may be minted on a private blockchain. A recipient may request the token by providing the unique token identifier and a zero-knowledge proof generated, by the recipient, based on the token secret. Based on comparing the zero-knowledge proof and the zero-knowledge proof data, the NFT may be sent to the recipient. A hash corresponding to a recipient address and the unique token identifier may be stored in a public blockchain.

Abstract: Aspects described herein may automatically generate threat models using large language model (LLM). A computing device may send, to LLM, one or more software modules associated with a computing system. The computing device may request the LLM to generate a threat model of the computing system. The computing device may receive, from the LLM, a first output based on the first prompt comprising first information for a first version of the threat model and a penetration test script for the computing system. The computing device may input, to the LLM, the result of the penetration test together with the LLM's previous output, to facilitate the LLM to generate a refined version of the threat model.

Abstract: Aspects described herein may automatically generate adversarial examples configured to train a machine learning model. A computing device may receive a request to generate a plurality of adversarial examples for a first machine learning model. The plurality of adversarial examples may be configured to be input to the first machine learning model and cause misclassification by the first machine learning model. The plurality of adversarial examples may be generated using a second machine learning model modified from a ground truth example. The computing device may send, to the first machine learning model, the plurality of adversarial examples. The first machine learning model may be configured to be adjusted based on a comparison between a respective output classification for each of the plurality of adversarial examples; and data indicating a correct classification for each of the plurality of adversarial examples.

Abstract: Systems and methods for executing domain-specific controls on large language model-generated data are disclosed herein. The system may receive a textual communication and provide the textual communication to a first model to generate an output. Based on the output and the textual communication, the system may generate a communication profile. The system may determine that the communication profile satisfies first or second criteria. Based on determining that the communication profile satisfies the first criteria, the system may determine rulesets corresponding to domains and provide the communication to a second model to generate a second output according to these rulesets. Based on determining that the communication profile satisfies the second criteria, the system may cause execution of a termination protocol in lieu of generating the second output.

Abstract: Methods, systems, and apparatuses are described herein for protecting the privacy of parties conducting Non-Fungible Token (NFT) transfers by conducting separate NFT transactions on a private blockchain separate from a public blockchain. An issuer and recipient may generate a token secret, and the issuer may send a create token transaction request comprising a unique token identifier, the token secret, and zero-knowledge proof data. Based on that request, an NFT may be minted on a private blockchain. A recipient may request the token by providing the unique token identifier and a zero-knowledge proof generated, by the recipient, based on the token secret. Based on comparing the zero-knowledge proof and the zero-knowledge proof data, the NFT may be sent to the recipient. A hash corresponding to a recipient address and the unique token identifier may be stored in a public blockchain.

Abstract: In some implementations, a device may obtain a set of biometric measurements, including a first type and a second type, at least one of the first type or the second type being a dynamic type. The device may evaluate the set of biometric measurements using a multi-modal artificial intelligence model, the multi-modal artificial intelligence model to generate an output prediction of a likelihood of the set of biometric measurements corresponding to stored characteristics of the single entity. The device may authenticate access for the single entity based on the output prediction from the multi-modal artificial intelligence model.

Abstract: Embodiments disclosed are directed to a computing system that performs steps to automatically identify risk control features and entities in a risk control document. The computing system uses a generative machine learning (ML) model to transform a risk control document into sequences of words, classify risk control features associated with the sequences of words, and pair the sequences of words with the classified risk control features. The computing system then uses a natural language processing (NLP) model to identify syntactic characteristics of the sequences of words. Subsequently, the computing system uses a discriminative predictor system to correct the classified risk control features based on the identified syntactic characteristics, identify boundaries of the corrected classified risk control features, and pair the identified boundaries with the corrected classified risk control features.

Abstract: Embodiments disclosed are directed to a computing system that performs steps to automatically identify risk control features and entities in a risk control document. The computing system uses a generative machine learning (ML) model to transform a risk control document into sequences of words, classify risk control features associated with the sequences of words, and pair the sequences of words with the classified risk control features. The computing system then uses a natural language processing (NLP) model to identify syntactic characteristics of the sequences of words. Subsequently, the computing system uses a discriminative predictor system to correct the classified risk control features based on the identified syntactic characteristics, identify boundaries of the corrected classified risk control features, and pair the identified boundaries with the corrected classified risk control features.

Abstract: Embodiments disclosed are directed to a computing system that performs steps to automatically identify risk control features and entities in a risk control document. The computing system regenerates, by a semantic prediction machine learning (ML) model, phrases in a risk control document. The computing system then classifies, by the semantic prediction ML model, risk control features associated with the regenerated phrases. Subsequently, the computing system corrects, by a discriminative natural language processing (NLP) model, the classified risk control features based on the phrases and the regenerated phrases.

Abstract: Embodiments disclosed are directed to a computing system that performs steps to automatically suggest a word, phrase, or entity to complete a sequence in a risk control document. The computing system classifies, by a generative machine learning (ML) model, risk control features associated with phrases in a risk control document. The computing system then generates, by the generative ML model and based on the classified risk control features, suggested words, phrases, or entities to complete a sequence following a cursor position in the risk control document. The computing system then corrects, by a discriminative natural language processing (NLP) model with domain specific knowledge, the suggested words, phrases, or entities. Subsequently, the computing system generates, by a discriminative predictor system, an encoded sequence of word, phrase, or entity suggestions based on the cursor position, the classified risk control features, and the corrected suggested words, phrases, or entities.