10068results about "Natural language translation" patented technology

A method is provided for estimating the number of distinct tokens in a text stream using a modified text-to-text variational autoencoder (T5VQVAE) model. The method includes receiving a continuous input of a text stream; dynamically maintaining a buffer that stores a probabilistic subset of tokens from the text stream; calculating a sampling probability for each token based on a condition related to the current state of the buffer; updating the buffer based on the sampling probability to include or exclude tokens; encoding the buffered tokens into a latent space using the T5VQVAE model; and estimating the number of distinct tokens in the text stream based on the tokens in the buffer and the corresponding sampling probabilities.

An example operation may include one or more of identifying a first topic from a call actively in progress, displaying a dashboard on a user device on the call, wherein the dashboard comprises content related to the first topic, receiving discussion data from the call, determining that a focus of the call has shifted from the first topic to a second topic, executing an artificial intelligence (AI) model on the second topic, dynamically generating dashboard content based on the execution, and displaying the dynamically generated dashboard content to the dashboard on the user device.

Methods and systems for retrieval-augmented generation are described. Responsive to a user input, an input embedding associated with the user input is obtained. A synthetic question embedding is retrieved from an embeddings database, based on a similarity to the input embedding. The synthetic question embedding is used to obtain a relevant source text based on a stored mapping between the synthetic question embedding and the source text. A prompt is provided to a large language model (LLM) to generate and display a textual response to the user input, based on the user input and the source text. The disclosed methods and systems effectively narrow the pool of source documents based on similarity measures between the user input embedding and the synthetic question embedding, to enable the retrieval of more relevant sources for use in response generation.

The invention relates to the technical field of operation and maintenance management and control, and discloses an intelligent building remote operation and maintenance management and control system based on a large model and a cloud edge collaborative architecture, and the system comprises a configuration subsystem which is used for carrying out the reverse proxy and persistent connection configuration of edge gateway equipment in an intelligent building, and obtaining a TLS encryption bidirectional communication channel; the construction subsystem is used for constructing a cloud-edge-end three-layer cooperative computing architecture; the conversion subsystem is used for performing unified semantic abstraction and conversion on the heterogeneous protocol data to obtain semantic data in a standard JSON (JavaScript Object Notation) format; the rule processing subsystem is used for performing rule processing through a near-end decision engine of the edge gateway to obtain a local intelligent response result; the multi-dimensional analysis subsystem is used for carrying out multi-dimensional analysis through a cloud large model engine to obtain an equipment health score, a fault prediction result and an energy optimization strategy, seamless fusion and interoperation of heterogeneous system data are achieved through the system, and then global optimization and linkage control of the intelligent building are achieved.

The present invention provides a large language model-based system and method for data processing in application developed by codeless platform. The invention includes identification of intent of a user to process procurement, supply chain, application integration, application restructuring or development scenarios.

In some embodiments, a language model forward traversal with a few-shot learning forward prompt yields a primary answer from a primary question. Then at least one backward traversal yields at least one candidate question using backward prompt(s) with answer-question pairs derived from the forward prompt's question-answer pairs. Each backward prompt also includes the primary answer but not the primary question. Each backward traversal is through one or more language models, not necessarily including the forward traversal's language model. Sometimes backward traversals vary model temperature, top-p, or top-k. A vector distance calculated between at least some candidate question vectors and a primary question vector indicates whether the primary answer includes hallucination content, and in some cases how much. Some embodiments withhold hallucinated answers from user interfaces and device control interfaces. Some embodiments also loop to obtain an answer with less hallucination content.

A database system generates a prompt for an LLM or other machine learning (ML) model to narrow the search space to highly relevant information about a database. A distinct instance of a classifier, a clustering algorithm, or a topic modeling model can be trained based on information from ML automation within the database system, respectively for each column or table in the database. Model instances can then be used during generative LLM inferencing to identify relevant sources of data to answer the user's question. Thus, the prompt generation combines ML automation and other ML models or an LLM for topic modeling and schema description.

An example may receive at least one input via at least one device. An example may use the at least one input to determine an entity identity. An example may use the entity identity to create an automated agent and load context data associated with the entity identity into at least one layer of a multi-layer memory of the automated agent. An example may cause the automated agent to machine-learn a supervision level via the context data. The machine-learned supervision level may indicate a level of supervision of the automated agent by an entity associated with the entity identity. An example may configure the automated agent to execute a task on behalf of the entity and in accordance with the machine-learned supervision level.

Systems and methods for resource-efficient retrieval of information using a generative AI model are disclosed. An input query requesting information from a set of documents is used in a prompt for a generative AI model to generate a search query to identify the documents relevant to the input query and their respective relevancy scores. The input query is used as an input another model to determine a depth score indicating a predicted number of documents needed to retrieve the information. Based on the depth score and the relevancy scores of the relevant documents, the system extracts grounding data from the identified relevant documents to generate an answer synthesis prompt for the generative AI model. The generative AI model processes the second to produce a response to the input query including the requested information.

Verifying large language model responses involves obtaining a query and its corresponding answer from a large language model. This conversational text is then fed into a second large language model, which translates the answer into first-order logic. The verification process uses an automated theorem prover. It checks the validity of this logic translation by determining the unsatisfiability of two scenarios: one where the negation of the logic translation and domain-specific logic formulas are combined, and another where the logic translation itself is combined with these formulas. Based on this analysis, the theorem prover ascertains whether the translated answer is valid, invalid, or neither. The final step is communicating this verification status through an appropriate output medium, such as a graphical user interface, a database, or a report, providing a structured and methodical approach to assessing the accuracy and reliability of language model responses.

The present invention relates to the technical field of knowledge graph completion tasks, and particularly relates to a knowledge graph link prediction method. The method comprises: using a precoding model to obtain an embedding layer vector, and constructing a corresponding masked triple; adding a corresponding position code to each element in the masked triple, so as to obtain a corresponding input sequence, inputting the input sequence into a trained main masking model, and outputting an entity classification probability; and on the basis of the entity classification probability, predicting potential candidate entities. The method further comprises: concatenating semantic information corresponding to the embedding layer vector and structural information obtained by an embedding model, so as to obtain fused head entity and relation representations, and constructing a corresponding fused masked triple; and adding a corresponding position code to each element in the fused masked triple, so as to obtain a corresponding fused input sequence. The present invention uses a precoding method, thereby effectively reducing the training burden on a model, and improving the inference speed of a model; and a fusion module is used before inputs are fed into a main masked model, thereby ensuring the integrity of textual description information and improving prediction accuracy.

Outputs generated by an LLM may depend significantly on the precise language of instructions and context included in a prompt into the LLM. However, can be difficult for users to define the exact language of instructions and context to be included in the prompt to generate desirable outputs. A computer-implemented method may involve: obtaining a candidate prompt including input data and instructions; inputting the candidate prompt into a LLM and receiving at least one candidate output generated by the LLM; receiving user input directed to the candidate prompt and / or the at least one candidate output; inputting at least the user input and one or both of the candidate prompt and the candidate output into the LLM as a modification prompt and receiving a subsequent candidate prompt generated by the LLM. The subsequent candidate prompt may include modified candidate instructions different from the candidate instructions.

Techniques are disclosed for enabling creators to create multimodal chatbots that are based on or simulate / model characters. The characters may be from audiovisual (AV) media such as films and TV shows or real people. The application leverages a combination of Visual Interpretation AI, Retrieval-Augmented Generation (RAG), Low-Rank Adaptation (LoRA), and function calling to provide rich, interactive experiences. The inferencing performed by an instant multimodal chatbot utilizes base weights, character weights, relationship weights, experience weights as we as environmental inputs. An instant chatbot uses a number of AI models including a video to text model, an image to text model, a sensory to text model, a large language model, a text to video model, a text to image model and a text to voice model. A user can interact with the chatbot in a variety of ways including text, audio and video.

A computer program comprising instructions which, when executed by a computer, cause the computer to perform the following operations: Receiving an input document (104), wherein the input document (104) contains content elements including at least one text element; Performing a lossless hierarchical text processing process (106) in which the input document (104) is reproduced as a hierarchical representation (108) in which all text elements from the input document (104) are preserved verbatim and organized according to a structure of the input document (104), the lossless hierarchical text processing process (106) comprising: Prompting a generative language model to generate the hierarchical representation (108), wherein the prompt instructs the generative language model to generate the hierarchical representation (108) without summarizing or omitting any of the text elements; and Performing a retrieval unit generation process (110) in which retrieval units (112) are generated based on the hierarchical representation (108) and stored in a memory accessible by a generative client language model for retrieval-enhanced generation.

Techniques for using a model to generate a response to a user input, where the response is associated with a personality determined to be relevant to the user input, are described. The system receives a user input and context data associated with the user input. Using the user input data and / or the context data, the system determines a personality (e.g., including a personality type and / or personality characteristics) relevant to the user input. The system generates a prompt instructing a model to generate a response to the user input that corresponds to the personality. The model processes the prompt to generate a response to the user input that corresponds to the personality. In some embodiments, the model generates a request for another component of the system to generate information responsive to the user input. The model may transform the responsive information into the personality-associated response.

Natural language queries can be answered by using a large language model to find an appropriate user interface element appearing in an application. User interface element metadata can be incorporated when choosing the user interface element. Browser automation can then navigate to a page in the application on which the interface element appears, extract answer data, and then present the answer data as an answer to the natural language query. Input values can be supported, and a large language model can select a matching input value based on semantic matching, even if the natural language query does not have an exactly matching input value. Additional features such as pre-calculating embeddings, pre-determining candidate input values, and the like can be supported. The technologies can provide natural language access to web applications that can result in immediate access for new users and faster query execution by experienced users.

The present invention provides a system and method for data processing in procurement and supply chain application developed by codeless platform. The invention includes one or more large language model (LLM) agents configured for processing one or more input received on an electronic user interface. The invention includes selecting a tool selection agent from a tool repository and invoking the selected tool by a tool execution agent for processing at least one task to be executed.

Disclosed in the present disclosure is a rich-media document auxiliary generation apparatus. The apparatus comprises a material extraction module, a theme sorting module, a semantic retrieval module, a structured data text generation module, an illustration recommendation module and a video composition module. The present disclosure uses intelligent means to assist a user to efficiently generate a high-quality rich-media composite document, thereby quickly and accurately describing a theme event in an all-round way.

According to an aspect of the present invention, there is provided an electrical audio signal processing system and device, comprising: a computer graphics processing and selective visual display system with a screen; an eye tracking device; a processor; one or more computer memory devices; wherein the processor is arranged for operations comprising: measuring the user's eye movements to ascertain the specific word on which the user is fixated, by the eye tracking device; modifying the display at the user's current fixation point; applying a delay between the presentation of successive graphic elements based on the user's calculated rate to accommodate the user's required time; and presenting elements to the user at a rate based upon the user's required time.

The disclosure relates to systems and methods of identifying concepts in content having natural language text using a Large Language Model (LLM), training neural networks in a discovery phase to classify the identified concepts into categories, sub-categories, or other groupings of concepts, and executing the neural networks in an operational phase to classify identified concepts.

Systems or techniques are provided for facilitating retrieval augmented generative question and answer boosting. In various embodiments, a system can access a plain text question regarding a scientific instrument. In various aspects, the system can generate, via a large language model that references a document-graph repository, a structured or unstructured answer for the plain text question. In various instances, the document-graph repository can comprise a plurality of document-graphs that respectively correspond to a plurality of technical documents. In various cases, for a first document-graph that corresponds to a first technical document, leaf nodes of the first document-graph can represent respective text blocks written in the first technical document, and non-leaf nodes of the first document-graph can respectively represent a document title, one or more section headings, and one or more scientific instrument identifiers written in the first technical document and beneath which the respective text blocks are nested.

A neural end-to-end system is provided for the face and voice preserving translation of videos. The system is a pipeline of multiple models that produces a video of the original speaker speaking in the target language with modified lip movement to match the target speech, while preserving emphases and prosody of the original speech, and voice characteristics of the original speaker. The pipeline starts with automatic speech recognition including emphasis detection, followed by the translation model. The translated text is then synthesized by a Text-to-Speech model that recreates the original emphases in the target sentence. The resulting synthetic speech is then converted back to the original speakers' voice using a voice conversion model. Finally, to synchronize the lips of the speaker with the translated audio, a generative model generates frames of adapted lip movements which are combined with the audio to produce the final output. The disclosure further describes several use-cases and configurations that apply these techniques to video conferencing, dubbing, low-bandwidth transmission, speech enhancement and assistive technology for the hearing impaired.

A computer-implemented method for operating an artificial intelligence voice agent system includes receiving voice input through communication channels; analyzing converted text through natural language processing (NLP) pipelines implementing intent recognition and sentiment analysis detecting emotional cues using a multimodal large language model (LLM); generating response content using machine learning models trained on domain-specific corpora; converting generated responses to synthetic speech through text-to-speech (TTS) engines; integrating with a customer relationship management (CRM) platforms or an enterprise resource planning (ERP) database; and implementing continuous learning by updating language understanding models using conversation logs, voice recognition parameters based on user feedback, and response generation patterns. One implementation is a computer-implemented system and method that operates a suite of intelligent interactive devices and platforms including an artificial intelligence voice agent, enhanced communication platforms, an intimacy companion system, and augmented / virtual reality eyeglasses. Further, one implementation includes AR / VR eyeglasses that project visual content onto interchangeable lenses or directly onto the user's retina via laser-based retinal projection, provide prescription adjustments, incorporate ear-mounted sensors for monitoring physiological parameters like heart rate, oxygen saturation, and blood pressure, and utilize wireless data transmission, onboard environmental sensing, and remote calibration, all designed to offer dynamically adaptive, secure, and context-aware interactions across communication, personal assistance, health monitoring, and immersive augmented or virtual reality environments.

The invention provides an evaluation method, system and device for a large language model and a storage medium, and the method comprises the steps: constructing a sample data set which comprises input data and corresponding reference answers, inputting the input data into an evaluated model, and generating an output result; a multi-dimensional evaluation index system is designed according to task requirements, a dynamic weight is allocated to each evaluation index, each evaluation index is provided with scoring standard description, and the evaluation indexes comprise at least two items of context correlation, term consistency, language fluency and expression accuracy; combining the input data, the reference answer, the output result and the scoring standard description into a standardized input instruction, and calling an evaluation model to perform multi-dimensional scoring on the standardized input instruction to generate an evaluation result; and analyzing the evaluation result according to a preset threshold value, and generating a structured feedback suggestion containing an improvement direction. According to the method, rapid optimization and iteration of the large language model can be effectively supported.

Techniques for updating a large language model (LLM) to correct generation of undesired responses, such as incorrect outputs, toxic outputs, etc. are described. Typical methods of retraining and fine-tuning are inefficient and computationally expensive for LLMs. Some embodiments of the present disclosure involve identifying a salient layer of the LLM that is responsible for the undesired response and editing only the salient layer. This layer is identified by computing a saliency value for the layer using a mean of gradient values for the layer, and the layer with the greatest saliency value is selected for editing. For editing, a small network is used to update the weights of the selected layer. The LLM is updated to include the edited layer, and the updated LLM is used for future processing.

The invention provides a minority language professional question and answer method based on mixed retrieval and retrieval enhancement generation, which comprises the following steps: S1, constructing a multi-language knowledge constructing a vector index database and a term knowledge graph by using a multi-language model according to a related minority language document; s2, multi-layer mixed retrieval: cross-language document recall is realized through a multi-layer mixed retrieval module, and the multi-layer mixed retrieval module is composed of keyword retrieval, semantic retrieval and vector retrieval; s3, answer generation: performing answer generation through an adaptive multi-language model by using a retrieval enhancement generation module, and introducing rule constraint decoding and a dynamic attention mechanism in the generation stage to improve the professionality and accuracy of the answer; and S4, self-adaptive optimization and knowledge updating: through a user feedback reinforcement learning module, optimizing the model based on user error correction data and supervising updating of the knowledge base. According to the method, professional questions and answers of the minority language can be realized based on a small amount of professional data of the minority language, the recall rate of the document of the minority language is improved, and the generation quality is improved.

An AI chatbot responds to the intent of a customer question by triggering an automatic workflow appropriate for the intention of the question. An information retrieval pipeline may be initiated to response to question corresponding to an information request. A Large Language Model may be initiated to generate a workflow to respond to other types of questions. The Large Language Model is provided with policies, tools and prompts to implement workflows. An evaluation engine evaluates factualness and helpfulness of responses to information questions and workflow intent accuracy and workflow appropriateness. Overall conversation resolution verification ay also be performed.

Disclosed are methods, systems, and non-transitory computer readable memory for processing, analyzing, and visualizing complex digital object sets using large language models. For instance, a method may include receiving a set of digital objects; indexing the received set of digital objects to generate indexed data; generating a timeline prompt based on the indexed data; processing the timeline prompt to generate a timeline response; and outputting the timeline response as an interactive timeline of based on the set of digital objects.

A multimodal assistant system receives a multimodal input that includes an input image and a natural language task description which describes a structured task to perform based on the input image. The system generates a visual Chain-of-Thought (v-CoT) prompt by combining the input image, the natural language task description, and a v-CoT instruction, the v-CoT instruction defining a series of steps for the MLLM to take in generating an output. The series of steps includes instructing the MLLM to describe relevant information derived from the image and relevant image artifacts required to generate the output, instructing the MLLM to reason over the relevant information and relevant image artifacts to determine a solution for the task, and including the solution in the output.

Verifying large language model responses involves obtaining a query and its corresponding answer from a large language model. This conversational text is then fed into a second large language model, which translates the answer into first-order logic. The verification process uses an automated theorem prover. It checks the validity of this logic translation by determining the unsatisfiability of two scenarios: one where the negation of the logic translation and domain-specific logic formulas are combined, and another where the logic translation itself is combined with these formulas. Based on this analysis, the theorem prover ascertains whether the translated answer is valid, invalid, or neither. The final step is communicating this verification status through an appropriate output medium, such as a graphical user interface, a database, or a report, providing a structured and methodical approach to assessing the accuracy and reliability of language model responses.