Abstract: Some implementations relate to generating a training and/or evaluation dataset with LLM prompts (e.g., derived from user queries) based on a prompt complexity. An input prompt, for example derived from a user query, is received. The input prompt is decomposed into a prompt tree comprising a plurality of nodes. The plurality of nodes comprise: a plurality of leaf nodes corresponding to simple sub-prompts of the input query; a plurality of branch nodes of sub-prompts each corresponding to multiple simple sub-prompts; and a root node corresponding to the input prompt. A prompt complexity is determined based on a path length of the prompt tree. The prompt complexity is compared to a threshold complexity. If the prompt complexity is above the threshold complexity, the input prompt is included in a set of training prompts and/or a set of evaluation prompts.

Abstract: Implementations relate to generating multi-modal response(s) through utilization of large language model(s) (LLM(s)). Processor(s) of a system can: receive natural language (NL) based input, generate a multi-modal response that is responsive to the NL based output, and cause the multi-modal response to be rendered. In some implementations, and in generating the multi-modal response, the processor(s) can process, using a LLM, LLM input (e.g., that includes at least the NL based input) to generate LLM output, and determine, based on the LLM output, textual content for inclusion in the multi-modal response and multimedia content for inclusion in the multi-modal response. In some implementations, the multimedia content can be obtained based on a multimedia content tag that is included in the LLM output and that is indicative of the multimedia content. In various implementations, the multimedia content can be interleaved between segments of the textual content.

Abstract: Implementations relate to a method implemented by one or more processors, the method including: receiving natural language (NL) based input associated with a client device; generating, using a large language model (LLM) and based on processing the NL based input, LLM output; determining, based on the LLM output, a sequence of LLM responses, the sequence of LLM responses including at least one intermediate LLM response and a final LLM response. In some implementations, the method may further include causing the final LLM response to be rendered at the client device. In additional or alternative implementations, the method may further include storing, as an instance of training data for fine-tuning the LLM or an additional LLM, the NL based input along with the final LLM response.

Abstract: Implementations relate to asymmetric quantization of large language models (LLMs). Processor(s) of a system can: obtain a trained LLM, wherein the trained LLM includes a plurality of layers, each layer comprising a respective plurality of weights; for each layer of the plurality of layers: calculate an optimal clipping range for the respective plurality of weights, and clip one or more weights of the respective plurality of weights that lie outside of the optimal clipping range to produce a clipped layer; quantize the LLM to generate a quantized LLM, wherein the instructions to quantize include instructions to map weights of the plurality of clipped layers of the LLM from continuous values to discrete values; and provide the quantized LLM for downstream processing.

Abstract: Implementations relate to reducing latency in generating and/or rendering a given stream of natural language (NL) based output generated using a large language model (LLM). Processor(s) of a system can: receive NL based input associated with a client device, generate the stream of NL based output utilizing the LLM that is responsive to the NL based input and that is for a given dialog context of an ongoing dialog, and cause the stream of NL based output to be rendered at the client device. Notably, the processor(s) can employ attribute classifier(s) and a multi-objective scorer to implement a blockwise controlled decoding technique in generating the stream of NL based output utilizing the LLM. By implementing the blockwise controlled decoding technique in generating the stream of NL based output utilizing the LLM, the processor(s) can reduce latency in generating and/or of the stream of NL based output generated utilizing the LLM.

Abstract: Implementations relate to generating multi-modal response(s) through utilization of large language model(s) (LLM(s)). Processor(s) of a system can: receive natural language (NL) based input, generate a multi-modal response that is responsive to the NL based output, and cause the multi-modal response to be rendered. In some implementations, and in generating the multi-modal response, the processor(s) can process, using a LLM, LLM input (e.g., that includes at least the NL based input) to generate LLM output, and determine, based on the LLM output, textual content for inclusion in the multi-modal response and multimedia content for inclusion in the multi-modal response. In some implementations, the multimedia content can be obtained based on a multimedia content tag that is included in the LLM output and that is indicative of the multimedia content. In various implementations, the multimedia content can be interleaved between segments of the textual content.

Abstract: Devices, computer-readable media, and methods for making a cache admission decision regarding a video chunk are described. For instance, a processing system including at least one processor may obtain a request for a first chunk of a first video, determine that the first chunk is not stored in a cache, and apply, in response to the determining that the first chunk is not stored in the cache, a classifier to predict whether the first chunk will be re-requested within a time horizon, where the classifier is trained in accordance with a set of features associated with a plurality of chunks of a plurality of videos. When it is predicted via the classifier that the first chunk will be re-requested within the time horizon, the processing system may store the first chunk in the cache.

Abstract: Provided are methodology and system countering fraudulent document and/or image use when authentication of a transaction based on a given document or image use is required. Additionally provided is a manner of machine learning adapting the methodology for implementation thereof.

Abstract: A method for regression and time series forecasting includes obtaining a set of hierarchical time series, each time series in the set of hierarchical time series including a plurality of time series data values. The method includes determining, using the set of hierarchical time series, a basis regularization of the set of hierarchical time series and an embedding regularization of the set of hierarchical time series. The method also includes training a model using the set of hierarchical time series and a loss function based on the basis regularization and the embedding regularization. The method includes forecasting, using the trained model and one of the time series in the set of hierarchical time series, an expected time series data value in the one of the time series.

Abstract: Devices, computer-readable media, and methods for making a cache admission decision regarding a video chunk are described. For instance, a processing system including at least one processor may obtain a request for a first chunk of a first video, determine that the first chunk is not stored in a cache, and apply, in response to the determining that the first chunk is not stored in the cache, a classifier to predict whether the first chunk will be re-requested within a time horizon, where the classifier is trained in accordance with a set of features associated with a plurality of chunks of a plurality of videos. When it is predicted via the classifier that the first chunk will be re-requested within the time horizon, the processing system may store the first chunk in the cache.

Abstract: A device, system and method for content-adaptive resolution-enhancement is provided. A plurality of subframe streams are generated from a video stream, each of the plurality of subframe streams comprising a lower resolution version of the video stream, pixel-shifted from one another. A plurality of output subframe streams are generated from the plurality of subframe streams in a one-to-one relationship by: applying a plurality of video enhancement filters to each of the plurality of subframe streams, each of the plurality of video enhancement filters for enhancing different features of the video stream; and, combining one or more resulting enhanced subframe streams into a respective output subframe stream based on data in one or more regions of the video stream. One or more projectors are controlled to project the plurality of output subframe streams to combine the plurality of output subframe streams into a higher resolution projected video stream.

Abstract: A device, system and method for content-adaptive resolution-enhancement is provided. A plurality of subframe streams are generated from a video stream, each of the plurality of subframe streams comprising a lower resolution version of the video stream, pixel-shifted from one another. A plurality of output subframe streams are generated from the plurality of subframe streams in a one-to-one relationship by: applying a plurality of video enhancement filters to each of the plurality of subframe streams, each of the plurality of video enhancement filters for enhancing different features of the video stream; and, combining one or more resulting enhanced subframe streams into a respective output subframe stream based on data in one or more regions of the video stream. One or more projectors are controlled to project the plurality of output subframe streams to combine the plurality of output subframe streams into a higher resolution projected video stream.