As an indispensable ingredient of intelligence, commonsense reasoning is crucial for large language models (LLMs) in real-world scenarios. In this paper, we propose CORECODE, a dataset that contains abundant commonsense knowledge manually annotated on dyadic dialogues, to evaluate the commonsense reasoning and commonsense conflict detection capabilities of Chinese LLMs. We categorize commonsense knowledge in everyday conversations into three dimensions: entity, event, and social interaction. For easy and consistent annotation, we standardize the form of commonsense knowledge annotation in open-domain dialogues as "domain: slot = value". A total of 9 domains and 37 slots are defined to capture diverse commonsense knowledge. With these pre-defined domains and slots, we collect 76,787 commonsense knowledge annotations from 19,700 dialogues through crowdsourcing. To evaluate and enhance the commonsense reasoning capability for LLMs on the curated dataset, we establish a series of dialogue-level reasoning and detection tasks, including commonsense knowledge filling, commonsense knowledge generation, commonsense conflict phrase detection, domain identification, slot identification, and event causal inference. A wide variety of existing open-source Chinese LLMs are evaluated with these tasks on our dataset. Experimental results demonstrate that these models are not competent to predict CORECODE's plentiful reasoning content, and even ChatGPT could only achieve 0.275 and 0.084 accuracy on the domain identification and slot identification tasks under the zero-shot setting. We release the data and codes of CORECODE at https://github.com/danshi777/CORECODE to promote commonsense reasoning evaluation and study of LLMs in the context of daily conversations.

Headline generation, a key task in abstractive summarization, strives to condense a full-length article into a succinct, single line of text. Notably, while contemporary encoder-decoder models excel based on the ROUGE metric, they often falter when it comes to the precise generation of numerals in headlines. We identify the lack of datasets providing fine-grained annotations for accurate numeral generation as a major roadblock. To address this, we introduce a new dataset, the NumHG, and provide over 27,000 annotated numeral-rich news articles for detailed investigation. Further, we evaluate five well-performing models from previous headline-generation tasks using human evaluation in terms of numerical accuracy, reasonableness, and readability. Our study reveals a need for improvement in numerical accuracy, demonstrating the potential of the NumHG dataset to drive progress in number-focused headline generation and stimulate further discussions in numeral-focused text generation.

Topic models have many important applications in fields such as Natural Language Processing. Topic embedding modelling aims at introducing word and topic embeddings into topic models to describe correlations between topics. Existing topic embedding methods use documents alone, which suffer from the topical fuzziness problem brought by the introduction of embeddings of semantic fuzzy words, e.g. polysemous words or some misleading academic terms. Links often exist between documents which form document networks. The use of links may alleviate this semantic fuzziness, but they are sparse and noisy which may meanwhile mislead topics. In this paper, we utilize community structure to solve these problems. It can not only alleviate the topical fuzziness of topic embeddings since communities are often believed to be topic related, but also can overcome the drawbacks brought by the sparsity and noise of networks (because community is a high-order network information). We give a new generative topic embedding model which incorporates documents (with topics) and network (with communities) together, and uses probability transition to describe the relationship between topics and communities to make it robust when topics and communities do not match. An efficient variational inference algorithm is then proposed to learn the model. We validate the superiority of our new approach on two tasks, document classifications and visualization of topic embeddings, respectively.

Accurate delineation of the volumetric motion of the left ventricle (LV) of the heart from tagged magnetic resonance imaging (MRI) is an important area of research. We have built a system that takes extracted tag line features from short axis (SA) and long axis (LA) image sequences as input and fits a four-dimensional (4-D) time-varying B-spline model to the data by simultaneously fitting the model knot solids to MRI frames via matching three sequences of solid knot planes to the LV tag planes for 4-D tracking. Important advantages of the model are that reconstruction of tag surfaces, three-dimensional (3-D) material point localization, as well as displacement reconstruction are all achieved in a single step. The generated 3-D displacement fields are validated with a cardiac motion simulator, and 3-D motion fields capturing in vivo deformations in a porcine model with posterolateral myocardial infarction are illustrated.

In this invited paper, an overview of techniques developed at the Cardiovascular Image Analysis Laboratory at Washington University is discussed. At the core of the authors' methodologies lie flexible shape models, which are employed in automated as well as semi-automated analysis of cardiac MRI and X-ray angiography images. The mathematical bases used for the flexible templates are of the B-spline variety, providing compact representation and interactive capabilities for manipulation of curves, surfaces, and volumes.

Accurate delineation of anatomical objects in 3D volumetric data is a significant problem in medical imaging. The authors have built a system that takes as input a spatial stack of 2D image slices being studied. The output of the system is a smooth 3D surface delineating the medical structure of interest, and the volume enclosed by the 3D surface. This paper presents a new method to represent a 3D anatomical object by a deformable B-spline tube, that facilitates the calculation of the volume enclosed based on a closed-form volume calculation formula. In order to validate the deformable B-spline surface fitting and associated volume calculation, 19 simulated 3D images were considered and theoretical volumes were compared with volumes produced by the authors' system. The errors were less than 4.9%.