Abstract: Embodiments described herein propose a densely connected Transformer architecture in which each Transformer layer takes advantages of all previous layers. Specifically, the input for each Transformer layer comes from the outputs of all its preceding layers; and the output information of each layer will be incorporated in all its subsequent layers. In this way, a L-layer Transformer network will have L(L+1)/2 connections. In this way, the dense connection allows the linguistic information learned by the lower layer to be directly propagated to all upper layers and encourages feature reuse throughout the network. Each layer is thus directly optimized from the loss function in the fashion of implicit deep supervision.

Abstract: A tag antenna comprises a feeder comprising a first conducting wire and a second conducting wire; a first electric conductor, a second electric conductor, and a third electric conductor that are elongated; a first connecting conductor; and a second connecting conductor. The first connecting conductor is separately connected to lower ends of the first electric conductor, the second electric conductor, and the third electric conductor; the second connecting conductor is connected to at least one of the first electric conductor, the second electric conductor, and the third electric conductor; and the first connecting conductor and the second connecting conductor are spaced apart, the first conducting wire of the feeder is connected to the first connecting conductor, and the second conducting wire is connected to the second connecting conductor. The embodiments of the present disclosure further provide a passive temperature detection apparatus.

Abstract: Systems and methods are provided that employ knowledge distillation under a multi-task learning setting. In some embodiments, the systems and methods are implemented with a larger teacher model and a smaller student model, each of which comprise one or more shared layers and a plurality of task layers for performing multiple tasks. During training of the teacher model, its shared layers are initialized, and then the teacher model is multi-task refined. The teacher model predicts teacher logits. During training of the student model, its shared layers are initialized. Knowledge distillation is employed to transfer knowledge from the teacher model to the student model by the student model updating its shared layers and task layers, for example, according to the teacher logits of the teacher model. Other features are also provided.

Abstract: Embodiments described herein propose a densely connected Transformer architecture in which each Transformer layer takes advantages of all previous layers. Specifically, the input for each Transformer layer comes from the outputs of all its preceding layers; and the output information of each layer will be incorporated in all its subsequent layers. In this way, a L-layer Transformer network will have L(L+1)/2 connections. In this way, the dense connection allows the linguistic information learned by the lower layer to be directly propagated to all upper layers and encourages feature reuse throughout the network. Each layer is thus directly optimized from the loss function in the fashion of implicit deep supervision.

Abstract: Dialogue summarization is challenging due to its multi-speaker standpoints, casual spoken language style, and limited labelled data. The embodiments are directed to a coarse-to-fine dialogue summarization model that improves abstractive dialogue summarization quality and enables granular controllability. A summary draft that includes key words for turns in a dialogue conversation history is created. The summary draft includes pseudo-labelled interrogative pronoun categories and noisy key phrases. The dialogue conversation history is divided into segments. A generate language model is trained to generate a segment summary for each dialogue segment using a portion of the summary draft that corresponds to at least one dialogue turn in the dialogue segment. A dialogue summary is generated using the generative language model trained using the summary draft.

Abstract: A tag antenna comprises a feeder comprising a first conducting wire and a second conducting wire; a first electric conductor, a second electric conductor, and a third electric conductor that are elongated; a first connecting conductor; and a second connecting conductor. The first connecting conductor is separately connected to lower ends of the first electric conductor, the second electric conductor, and the third electric conductor; the second connecting conductor is connected to at least one of the first electric conductor, the second electric conductor, and the third electric conductor; and the first connecting conductor and the second connecting conductor are spaced apart, the first conducting wire of the feeder is connected to the first connecting conductor, and the second conducting wire is connected to the second connecting conductor. The embodiments of the present disclosure further provide a passive temperature detection apparatus.

Abstract: Systems and methods are provided that employ knowledge distillation under a multi-task learning setting. In some embodiments, the systems and methods are implemented with a larger teacher model and a smaller student model, each of which comprise one or more shared layers and a plurality of task layers for performing multiple tasks. During training of the teacher model, its shared layers are initialized, and then the teacher model is multi-task refined. The teacher model predicts teacher logits. During training of the student model, its shared layers are initialized. Knowledge distillation is employed to transfer knowledge from the teacher model to the student model by the student model updating its shared layers and task layers, for example, according to the teacher logits of the teacher model. Other features are also provided.

Abstract: A system for detecting temperature during roasting includes a roasting probe and a transmission terminal. The roasting probe includes a temperature sensor and a first wireless transmission component. The temperature sensor is used to detect temperature of the roasting probe. The first wireless transmission component sends the detected temperature to the transmission terminal. The transmission terminal includes a shell, a circuit board arranged in the shell, and a second wireless transmission component and a third wireless transmission component connected to the circuit board. When the transmission terminal is connected to a roasting apparatus, the second wireless transmission component enters the roasting apparatus and is used to receive the temperature sent by the first wireless transmission component located on the roasting probe in the roasting apparatus. The third wireless transmission component is located outside the roasting apparatus and sends the temperature to an external terminal.