Retrieval-Generation Alignment for End-to-End Task-Oriented Dialogue System

Developing an efficient retriever to retrieve knowledge from a large-scale knowledge base (KB) is critical for task-oriented dialogue systems to effectively handle localized and specialized tasks. However, widely used generative models such as T5 and ChatGPT often struggle to differentiate subtle differences among the retrieved KB records when generating responses, resulting in suboptimal quality of generated responses. In this paper, we propose the application of maximal marginal likelihood to train a perceptive retriever by utilizing signals from response generation for supervision. In addition, our approach goes beyond considering solely retrieved entities and incorporates various meta knowledge to guide the generator, thus improving the utilization of knowledge. We evaluate our approach on three task-oriented dialogue datasets using T5 and ChatGPT as the backbone models. The results demonstrate that when combined with meta knowledge, the response generator can effectively leverage high-quality knowledge records from the retriever and enhance the quality of generated responses. The codes and models of this paper are available at https://github.com/shenwzh3/MK-TOD.

Deeply supervised neural network with short connections for retinal vessel segmentation

The condition of vessel of the human eye is a fundamental factor for the diagnosis of ophthalmological diseases. Vessel segmentation in fundus image is a challenging task due to low contrast, the presence of microaneurysms and hemorrhages. In this paper, we present a multi-scale and multi-level deeply supervised convolutional neural network with short connections for vessel segmentation. We use short connections to transfer semantic information between side-output layers. Forward short connections could pass low level semantic information to high level and backward short connections could pass much structural information to low level. In addition, we propose using a structural similarity measurement to evaluate the vessel map. The proposed method is verified on DRIVE dataset and it shows superior performance compared with other state-of-the-art methods. Specially, with patch level input, the network gets 0.7890 sensitivity, 0.9803 specificity and 0.9802 AUC. Code will be made available at https://github.com/guomugong/sdsn.