Abstract: A method, apparatus, non-transitory computer readable medium, and system for document retrieval include obtaining a query and a document. A prompt generator generates a prompt for a reasoning model based on the query and the document. The reasoning model generates a reasoning result based on the prompt. In some cases, the reasoning result indicates that the document answers the query. A machine learning model provides the document in response to the query based on the reasoning result.

Abstract: Methods for determining optimal cloud service resource include determining a reward function for a set of resource configurations identifying cloud service resource parameters. The cloud service resource parameters include a source parameter and a target parameter of services to provide a client computing device. A source parameter dataset for the source parameter and a target parameter dataset is generated using the reward function and historical source parameter data. The matrices are then subject to SVD and clustering. A target parameter reward dataset is learned from output of the SVD and clustering. The target parameter dataset is used to determine the parameters for the target parameter for providing corresponding cloud service resources.

Abstract: Methods and systems are provided for using generative artificial intelligence to evaluate fine-tuned language models. In embodiments described herein, natural language text snippets are generated via a generative language model based on corresponding data. A language model is fine-tuned into a fine-tuned language model via a language model fine-tuning component using the natural language text snippets and the corresponding data as training data. Independent natural language text snippets are generated via the generative language model based on the corresponding data. Each independent natural language text snippet is different than each corresponding natural language text snippet. An evaluation metric of the fine-tuned language model is generated via an evaluation component based on the independent natural language text snippets and the corresponding data.

Abstract: A photovoltaic device and a method for mounting a photovoltaic device are provided. The photovoltaic device includes a color steel tile, a bonding layer, and a photovoltaic assembly. The color steel tile includes an angle relaxation portion. The bonding layer is arranged on a top wall of the angle relaxation portion. The photovoltaic assembly is located on one side of the color steel tile and connected to the color steel tile through the bonding layer. The bonding layer includes a first bonding portion and a second bonding portion arranged along a first direction. In the first direction, a ratio of a dimension of the first bonding portion to a dimension of the second bonding portion ranges from 0.1 to 0.5. The first bonding portion temporarily fix the photovoltaic assembly, while the second bonding portion mainly fix the photovoltaic assembly.

Abstract: Graphic visualizations, such as charts or graphs conveying data attribute values, can be generated based on natural language queries, i.e., natural language requests. To do so, a natural language request is parsed into n-grams, and from the n-grams, word embeddings are determined using a natural language model. Data attributes for the graphic visualization are discovered in the vector space from the word embeddings. The type of graphic visualization can be determined based on a request intent, which is determined using a trained intent classifier. The graphic visualization is generated to include the data attribute values of the discovered data attributes, and in accordance with the graphic visualization type.

Abstract: This disclosure describes one or more implementations of systems, non-transitory computer-readable media, and methods that train a named entity recognition (NER) model with noisy training data through a self-cleaning discriminator model. For example, the disclosed systems utilize a self-cleaning guided denoising framework to improve NER learning on noisy training data via a guidance training set. In one or more implementations, the disclosed systems utilize, within the denoising framework, an auxiliary discriminator model to correct noise in the noisy training data while training an NER model through the noisy training data. For example, while training the NER model to predict labels from the noisy training data, the disclosed systems utilize a discriminator model to detect noisy NER labels and reweight the noisy NER labels provided for training in the NER model.

Abstract: The present application relates to the technical field of computers. Provided is a small sample fine-turning method, the method comprising: inputting a data set, and forming an input sample according to a fixed template; constructing a candidate tag word set and a candidate prompt template set; by means of reinforcement learning, searching an optimal tag word corresponding to the input sample from the candidate tag word set, and a prompt template corresponding to the input sample from the candidate prompt template set; and outputting a mapping relationship of the optimal tag word and an optimal prompt template format corresponding to the prompt template.

Abstract: Graphic visualizations, such as charts or graphs conveying data attribute values, can be generated based on natural language queries, i.e., natural language requests. To do so, a natural language request is parsed into n-grams, and from the n-grams, word embeddings are determined using a natural language model. Data attributes for the graphic visualization are discovered in the vector space from the word embeddings. The type of graphic visualization can be determined based on a request intent, which is determined using a trained intent classifier. The graphic visualization is generated to include the data attribute values of the discovered data attributes, and in accordance with the graphic visualization type.

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods that implement a neural network framework for interactive multi-round image generation from natural language inputs. Specifically, the disclosed systems provide an intelligent framework (i.e., a text-based interactive image generation model) that facilitates a multi-round image generation and editing workflow that comports with arbitrary input text and synchronous interaction. In particular embodiments, the disclosed systems utilize natural language feedback for conditioning a generative neural network that performs text-to-image generation and text-guided image modification. For example, the disclosed systems utilize a trained model to inject textual features from natural language feedback into a unified joint embedding space for generating text-informed style vectors. In turn, the disclosed systems can generate an image with semantically meaningful features that map to the natural language feedback.

Abstract: One or more aspects of the method, apparatus, and non-transitory computer readable medium include obtaining image data and computing a prediction residue value for a pixel of the image data using a prediction function. An entropy value for the pixel can then be determined based on the prediction residue value using context modeling, and progressive compressed image data for the image data can be generated based on the entropy value. The compressed image data can be used to enable collaborative image editing and other image processing tasks.

Abstract: Dialogue skeleton assisted prompt transfer for dialogue summarization techniques are described that support training of a language model to perform dialogue summarization in a few-shot scenario. A processing device, for instance, receives a training dataset that includes training dialogues. The processing device then generates dialogue skeletons based on the training dialogues using one or more perturbation-based probes. The processing device trains a language model using prompt transfer between a source task, e.g., dialogue state tracking, and a target task, e.g., dialogue summarization, using the dialogue skeletons as supervision. The processing device then receives an input dialogue and uses the trained language model to generate a summary of the input dialogue.

Abstract: Systems and methods for reducing inference time of vision-language models, as well as for multimodal search, are described herein. Embodiments are configured to obtain an embedding neural network. The embedding neural network is pretrained to embed inputs from a plurality of modalities into a multimodal embedding space. Embodiments are further configured to perform a first progressive pruning stage, where the first progressive pruning stage includes a first pruning of the embedding neural network and a first fine-tuning of the embedding neural network. Embodiments then perform a second progressive pruning stage based on an output of the first progressive pruning stage, where the second progressive pruning stage includes a second pruning of the embedding neural network and a second fine-tuning of the embedding neural network.

Abstract: Methods for determining optimal cloud service resource include determining a reward function for a set of resource configurations identifying cloud service resource parameters. The cloud service resource parameters include a source parameter and a target parameter of services to provide a client computing device. A source parameter dataset for the source parameter and a target parameter dataset is generated using the reward function and historical source parameter data. The matrices are then subject to SVD and clustering. A target parameter reward dataset is learned from output of the SVD and clustering. The target parameter dataset is used to determine the parameters for the target parameter for providing corresponding cloud service resources.

Abstract: Dialogue state aware dialogue summarization techniques are described that enable generation of dialogue summaries from target domains with limited training data. A content processing system, for instance, generates one or more clusters based on training dialogues from one or more source domains. The clusters represent domain-specific features of the training dialogues and are further based on dialogue states of the training dialogues. The content processing system trains a machine learning model to generate summaries of dialogues by using the one or more clusters as prefixes in a prefix-tuning approach. The content processing system receives an input that includes a dialogue from a target domain. The content processing system generates an input prompt based on the dialogue and the one or more clusters, and the model generates a summary of the dialogue based on the input prompt.

Abstract: Contextual query generation techniques are described that enable generation of a contextual query for output to a question-answering (QA) model. A content processing system, for instance, configures a language model using in-context learning to generate queries based on semantic contexts of input documents, e.g., based on one or more linguistic cues from text of the input documents. The content processing system receives an input that includes a document having text and a reference query. The content processing system leverages the language model to generate a contextual query based on a semantic context of the text of the document and the reference query. The content processing system then outputs the contextual query and the document to a QA model. Using the QA model, the content processing system generates a response as an answer to the contextual query based on the contextual query and the document.

Abstract: A photovoltaic component, including a plurality of color steel tiles and a plurality of photovoltaic modules. Cavities are formed between the photovoltaic modules and the color steel tiles. The photovoltaic component further includes cables connecting two adjacent photovoltaic modules. The cables are provided with fixing portions. The fixing portions are configured to fix the cables to a side of the photovoltaic modules close to the color steel tiles. The cables are located in the cavities. Along a thickness direction of the photovoltaic modules, the cables and the color steel tiles are not in contact with each other.

Abstract: A negative electrode sheet includes a negative-electrode current collector and a negative-electrode material layer disposed on the negative-electrode current collector. The negative-electrode material layer includes a first active material that includes a first particle cluster and a second particle cluster. A compaction density of the first particle cluster after being compressed at a pressure of 5 tons is P1, and a compaction density of the second particle cluster after being compressed at a pressure of 5 tons is P2, where P1 and P2 satisfy: 1.0?P1/P2?1.5. When a volume percentage in the first particle cluster reaches 50%, a corresponding particle size value Dv50 satisfies: 13 ?m?Dv50?20 ?m, and when a volume percentage in the second particle cluster reaches 50%, a corresponding particle size value Dv50? satisfies: 5 ?m?Dv50??12 ?m.

Abstract: Systems and methods for multimodal machine learning are provided. According to one aspect, a method for multimodal machine learning includes obtaining a prompt; encoding the prompt using a multimodal encoder to obtain a prompt embedding, wherein the encoding comprises generating a plurality of multi-head attention (MHA) outputs corresponding to a plurality of different scales, respectively, and combining the plurality of MHA outputs using a multi-scale aggregator; and generating a response to the prompt based on the prompt embedding.

Abstract: Provided are a photovoltaic facility and a method for assembling a frame of a photovoltaic module. The photovoltaic facility includes: a clamp, a color steel tile, a photovoltaic module, and a connecting member. The clamp is connected to the color steel tile and includes a clamping body. The photovoltaic module is located on one side of the clamping body and is connected to the clamping body through the connecting member. The photovoltaic module includes a laminate and a frame. The frame is connected to a back side of the laminate and has an opening. The connecting member extends into the opening and abuts against a sidewall of the opening. The photovoltaic module is connected to the color steel tile through the clamp.

Abstract: Provided are a photovoltaic module and a photovoltaic facility. The photovoltaic module includes a laminate and a frame. The frame includes a first portion, a second portion, and a third portion. The second portion and the third portion are respectively connected to two ends of the first portion. The second portion and the third portion extend along a first direction, and the second portion is fixedly connected to a back surface of the laminate, or the second portion and the third portion extend along a third direction, and the first portion is fixedly connected to the back surface of the laminate. The frame supports the laminate.