Abstract: Model training systems collaborate on model training without revealing respective private data sets. Each private data set learns a set of client weights for a set of computer models that are also learned during training. Inference for a particular private data set is determined as a mixture of the computer model parameters according to the client weights. During training, at each iteration, the client weights are updated in one step based on how well sampled models represent the private data set. In another step, gradients are determined for each sampled model and may be weighed according to the client weight for that model, relatively increasing the gradient contribution of a private data set for model parameters that correspond more highly to that private data set.

Abstract: Systems and methods for retrieving information from a corpus of documents that is relevant to a query. The method comprising: subdividing each document in the corpus of documents into one or more chunks; using a synthetic generation LLM to generate at least one piece of synthetic information related to each chunk; generating, using an embedding model a plurality of vectors for each chunk, the plurality of vectors for a chunk comprising a vector generated from the chunk, and a vector generated from each of the at least one piece of synthetic information related to that chunk; and using an information retrieval system to identify, from the plurality of vectors for each chunk, a set of chunks that are relevant to a query.

Abstract: Approaches that intend to reduce disparate impact, particularly those for providing equal coverage sets in conformal prediction, can in fact increase disparate impact for human-in-the-loop systems. To improve these systems, rather than optimizing selection processes for a class prediction set for a confidence level (e.g., a percentage confidence that the correct class is in the class prediction set), a selection process is determined that reduces the set size difference across groups.

Abstract: Meta-learning models are improved for few-shot learning of unseen tasks by improving task diversity of training data used for training the meta-learning model. A task diversity score may be determined between a pair of tasks that partition a domain into respective classes. The respective classes are paired and scored to determine similarity between class pairs and subsequent task diversity scores. Diverse tasks may be generated with unsupervised analysis of the domain by determining disentangled latent features of the data samples. Each latent feature may then be considered a task with classes based on a clustering of the data samples based on the feature values of the respective latent feature. The classes are then used as training task labels for the data samples and sampled from to generate diverse tasks for the meta-learning model.

Abstract: Meta-learning models are improved for few-shot learning of unseen tasks by improving task diversity of training data used for training the meta-learning model. A task diversity score may be determined between a pair of tasks that partition a domain into respective classes. The respective classes are paired and scored to determine similarity between class pairs and subsequent task diversity scores. Diverse tasks may be generated with unsupervised analysis of the domain by determining disentangled latent features of the data samples. Each latent feature may then be considered a task with classes based on a clustering of the data samples based on the feature values of the respective latent feature. The classes are then used as training task labels for the data samples and sampled from to generate diverse tasks for the meta-learning model.

Abstract: Systems and method for generating a response to a query. The method includes using a first large language model (LLM) to generate synthetic information related to a query; generating an amended query based on the synthetic information related to the query; using an information retrieval system to retrieve, from a plurality of chunks, a set of chunks that are relevant to the amended query, wherein each chunk of the plurality of chunks is all or a portion of a document in a corpus of documents; using a second LLM to rank the set of chunks based on a relevance to the query; selecting a subset of chunks from the set of chunks based on the ranking; and using a third LLM to generate a response to the query based on the subset of chunks.

Abstract: Methods and systems for generating a response to a query. The methods comprising: receiving a set of chunks that are relevant to the query, the set of chunks from a plurality of chunks generated from documents in a corpus of documents; using a re-ranker LLM to rank the set of chunks that are relevant to the query based on a relevance to the query via chain-of-thought prompting; selecting a subset of chunks from the set of chunks based on the ranking; and using a generation LLM to generate the response to the query based on the subset of chunks.

Abstract: Systems and methods for generating a response to a query based on a corpus of documents. The method comprising: subdividing each document into one or more chunks; using an LLM to generate synthetic information related to each chunk; generating, using an embedding model, a plurality of vectors for each chunk, the plurality of vectors for a chunk comprising a vector generated from the chunk, and a vector generated from the related synthetic information; using an information retrieval system to identify, from the plurality of vectors for each chunk, a set of chunks that are relevant to a query; using an LLM to rank the set of chunks based on their relevance to the query via chain-of-thought prompting; selecting a subset of chunks from the set of chunks based on the ranking; using an LLM to generate a response to the query based on the subset of chunks.

Abstract: Systems and methods for retrieving information from a corpus of documents that is relevant to a query. The method comprises: generating a first plurality of chunks by subdividing each document in the corpus of documents into one or more chunks of a first size; generating a second plurality of chunks by subdividing each document in the corpus of documents into one or more chunks of a second, larger, size; using an information retrieval system to identify, from the second plurality of chunks, a set of chunks of the second size that are relevant to a query; and using the information retrieval system to identify, from a subset of chunks of the first plurality of chunks, a set of chunks of the first size that are relevant to the query. The subset is based on the set of chunks of the second size that are relevant to the query.

Abstract: Systems and methods for retrieving information from a corpus of documents that is relevant to a query. The method comprising: subdividing each document in the corpus of documents into one or more chunks; using a synthetic generation LLM to generate at least one piece of synthetic information related to each chunk; generating, using an embedding model a plurality of vectors for each chunk, the plurality of vectors for a chunk comprising a vector generated from the chunk, and a vector generated from each of the at least one piece of synthetic information related to that chunk; and using an information retrieval system to identify, from the plurality of vectors for each chunk, a set of chunks that are relevant to a query.

Abstract: Systems and methods for retrieving information from a corpus of documents that is relevant to a query. The method comprising: subdividing each document in the corpus of documents into one or more chunks; using a synthetic generation LLM to generate at least one piece of synthetic information related to each chunk; generating, using an embedding model a plurality of vectors for each chunk, the plurality of vectors for a chunk comprising a vector generated from the chunk, and a vector generated from each of the at least one piece of synthetic information related to that chunk; and using an information retrieval system to identify, from the plurality of vectors for each chunk, a set of chunks that are relevant to a query.

Abstract: A computer model (e.g., an artificial intelligence model) having an encoder that generates embeddings may undesirably generate relatively large differences in embeddings for small changes in the input data sample. To improve robustness of the model against this type of change, training samples may be modified to generate adversarial examples that have comparatively large embedding differences relative to the change in training data sample. The adversarial data samples may be generated iteratively by exploring perturbations of the training data sample within a threshold to increase the distance in the embedding space. A robust encoder for the model may then be trained with the training data sample and adversarial data sample to reduce the distance between the corresponding training embedding and adversarial embedding.

Abstract: Systems and methods for retrieving relevant documents. A computing system obtains, from each document in a corpus of documents, a plurality of chunks corresponding to portions of text. It computes a score for each one of the plurality of chunks in relation to a query. The chunks are reordered according to score. A sum of the highest scores is computed, and a subset of chunks associated with the highest scoring documents are retrieved. A large language model (LLM) may be used to generate response text from the retrieved documents.

Abstract: A sensor adjustment method includes obtaining a folding status parameter, where the folding status parameter is used to describe a folding extent of a foldable device, and where a component of the foldable device in a folded state causes interference to a first sensor of the foldable device compared with a non-folded state; invoking a sensor adjustment corresponding to the folding status parameter; and adjusting a sensor operating parameter of the first sensor according to the sensor adjustment and/or performing data preprocessing on first sensor data collected by the first sensor.

Abstract: A transmit power adjustment method, apparatus, and a system. The system is configured for obtaining a first distance between a first antenna and a second antenna, determining a first transmit power of the first antenna, and a second transmit power of the second antenna, wherein, when the first distance is greater than a first value, the first transmit power is at or below a first limit power, and the second transmit power is at or below a second limit power, or wherein, when the first distance is less than the first value, a sum of a first exposure ratio of the first transmit power and a second exposure ratio of the second transmit power is less than or equal to 1, and transmitting, a signal at the first transmit power through the first antenna and a signal at the second transmit power through the second antenna.

Abstract: Disclosed is a method for gene expression by transient transformation of a wild rice seed using Agrobacterium. The method includes the following steps: subjecting a wild rice seed to induction culture and subculture in sequence, and selecting a resulting callus with a dense structure to allow pre-culture to obtain a pre-cultured callus; transforming a 1305Ubi-Ubi-GFP-H plasmid carrying a green fluorescent protein (GFP) gene into an Agrobacterium strain EHA105 or LBA4404 to obtain a positive Agrobacterium strain; subjecting the positive Agrobacterium strain to culture to obtain a bacterial suspension with an optical density (OD) value of 0.02 to 0.5 at a wavelength of 600 nm to obtain an infection bacterial solution; and infecting the pre-cultured callus with the infection bacterial solution to allow co-culture and recovery culture in sequence; observing an infection status of a recovered callus under a stereo fluorescence microscope, and calculating a transient expression rate.

Abstract: Described here are systems and methods for a robust magnetic resonance elastography (“MRE”) imaging platform for rapid dynamic 3D MRE imaging. The imaging platform includes an MRE pulse sequence and advanced image reconstruction framework that work synergistically in order to greatly expand the domains where MRE can be deployed successfully.

Abstract: A method includes obtaining an average radio frequency exposure value over a target time window, where the target time window is a standard time window, determining a power threshold corresponding to the average radio frequency exposure value, where the power threshold is a preset power value corresponding to a preset value range to which the average radio frequency exposure value belongs, and controlling a transmit power of the wireless device based on the power threshold over a controlled time window.

Abstract: A variety of generative models are trained that are trained on a reference data set. The generative models are evaluated by candidate metrics to determine the relative rankings of the models as evaluated by the different candidate metrics. Rankings as generated by the models is compared with human evaluation of the generated results as simulated and the candidate metrics that most align with the human evaluation may then be used to automatically evaluate subsequent generative models. The candidate metrics may include various types of encoding models trained for non-generative purposes, such that the selected candidate metric may represent selecting an encoding model that performs well on the generative data.

Abstract: Described here are systems and methods for a robust magnetic resonance elastography (“MRE”) imaging platform for rapid dynamic 3D MRE imaging. The imaging platform includes an MRE pulse sequence and advanced image reconstruction framework that work synergistically in order to greatly expand the domains where MRE can be deployed successfully.