Provided are a server, method, and computer-readable recording medium for searching for a vector. The method includes generating a vector index structure of data points, searching for a node similar to a query vector using the vector index structure, calculating a similarity between the node and the query vector, and updating the similarity by giving a weight to a vector index inflow time of the node.

A user interaction command is obtained via an interface interaction mechanism associated with a viewer interface element structured to display page data associated with a set of page nodes that satisfy filter parameters of a view. A virtual scroll space position associated with the interface interaction mechanism resulting from the user interaction command is determined. A virtual pixel space data structure is queried to determine a virtual pixel space position corresponding to the virtual scroll space position. A determination is made that a buffer update threshold has been exceeded. The virtual pixel space data structure is queried to determine a buffered subset of page nodes based on: the virtual pixel space position, pixel ranges associated with the set of page nodes, and a height bound associated with the viewer interface element. Page data corresponding to the buffered subset of page nodes is loaded into the viewer interface element.

Embodiments are directed towards previewing results generated from indexing data raw data before the corresponding index data is added to an index store. Raw data may be received from a preview data source. After an initial set of configuration information may be established, the preview data may be submitted to an index processing pipeline. A previewing application may generate preview results based on the preview index data and the configuration information. The preview results may enable previewing how the data is being processed by the indexing application. If the preview results are not acceptable, the configuration information may be modified. The preview application enables modification of the configuration information until the generated preview results may be acceptable. If the configuration information is acceptable, the preview data may be processed and indexed in one or more index stores.

Systems, methods, and non-transitory computer-readable media are provided for conducting user query searches. According to one implementation, a process includes a step of, in response to receiving a user query directed to subject information retrievable from documentation stored in a private database, using a section-based chunking procedure to obtain, from the private database, a relevant section of the documentation as context. The process further includes a step of feeding the user query and the relevant section as context to a Large Language Model (LLM).

An application extracts a plurality of features of a hardware component of an aircraft. The application inputs a first subset of features of the plurality of features into a first machine learning model, and receives as output a first determination of whether the hardware component is rotable. The application inputs a second subset of features of the plurality of features into a second machine learning model, and receives as output a second determination of whether the hardware component is rotable. The applications determines, based on the first determination and the second determination, a final determination of whether the hardware component is rotable, and adds a data structure for the hardware component with the final determination in a searchable database. The application receives a query from a user that is associated with the hardware component, runs a search, outputs whether the hardware component is rotable.

Techniques for a provider network user console with a natural language querying feature. The techniques include collecting a set of templatized query pairs. A pair includes a templatized natural language query (NLQ) and a templatized domain-specific index query language query. The set of templatized query pairs is expanded by substituting named variable tokens in the templatized query pairs with synthetic values to create a set of instantiated query pairs. A pre-trained neural machine translation model is retrained using the set of instantiated query pairs to yield a fine-tuned neural machine translation model. For a natural language query received, the fine-tuned neural machine translation model is used to translate the target natural language query to a corresponding domain-specific index query language query. The domain-specific index query language query is then executed against an index to yield an index result. The techniques reduce or eliminate the need for manual training data generation.

Systems and methods for using large language models (LLM) to generate and edit content items are described. The methods generate a template of a content item based on user input and populate the template based on data obtained from querying an LLM generated semantic graph that includes data sources to which a user is authorized access. Once generated, on or more sources provided are used for editing the identified text of the generated content item. If multiple sources are provided, they are weighted based on factors and their priority is determined. Raw data from the provided source(s) is indexed. An LLM using the identified text as an input, queries the index to identify data relevant for editing the identified text. The relevant data is then used to edit, rewrite and/or regenerate the identified text and make the other unedited text in the content item coherent with the edited text.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for recurrence in a transformer architecture. In one aspect, a method includes receiving input embeddings representing a sequence of words as input; generating as output attention vectors for each of the words, the attention vectors for each word indicating an importance of the word in the sequence relative to other words in the sequence; generating first and second linear transformations X.sub.1 and X.sub.2 of the attention vectors; determining, in a recurrent neural network, a hidden state corresponding to each attention vector using only element wise operations on the first linear transformation of the attention vectors during a recurrent step; and generating a set of output vectors using a multiplicative gating function in combination with the second linear transformation.

Embodiments sort a plurality of documents in an increasing order of frequency updates, create multiple recursive partitions of the plurality of sorted documents, compute a cost of the multiple recursive partitions, choose a partition with a smallest cost from computed costs of the multiple recursive partitions, and merge the plurality of documents based on the partition with the smallest cost.

A method and related system may traverse a hierarchical data structure to dynamically generate text outputs based on user prompts. The method includes retrieving sub-data structures from multiple layers via links, applying a duration protocol and time constraints, and using a language model API to generate candidate output content. The method may also include determining whether the content meets time criteria and, if so, updating the first sub-data structure for display at a second client node.