As the popularity of AI-powered chatbots has grown, research on supportive communication has begun to explore chatbots' ability to provide social support. This two-study project examined if humans and ChatGPT detect linguistic variations in support-seeking messages (levels of self-disclosure, Study 1; linguistic agency and causality, Study 2) and how message variations affect quality of support provided. ChatGPT responses were higher in interaction involvement and politeness, but human responses demonstrated more reciprocal self-disclosure. In Study 1, depth of self-disclosure influenced quality of human responses, but not ChatGPT responses. In Study 2, human responses were relatively consistent across variations in linguistic agency and causality, but ChatGPT responses demonstrated greater cognitive and emotional involvement to messages identifying non-biological causes of depression. Our findings suggest that ChatGPT can provide quality support for various stressors, but it is less able to adapt to nuances of how users describe their experiences compared to human support providers.

Cross-prompt automated essay scoring (AES) aims to train models using essays from different source prompts and test them on new target prompt essays. A core challenge of the task is to learn as much shared knowledge as possible between essays from different prompts in order to better represent new prompt essays. Previous studies primarily focus on learning this knowledge on a general, coarse-grained level, ignoring that the shared knowledge among prompts is highly detailed and contains a more comprehensive range of information that is not fully investigated. In this paper, we propose a novel multi-aspect knowledge finding and aligning optimization strategy to better acquire this detailed various shared knowledge. We also introduce LLM to extract explicit, interpretable knowledge from implicit, multi-aspect shared knowledge and use this knowledge to improve the representation and evaluation performance of new prompt essays. We conduct extensive experiments on public datasets. The results show that our approach outperforms current state-of-the-art models and is effective on cross-prompt AES.

Most current automated essay quality assessment systems treat score prediction and feedback generation as separate tasks, overlooking the fact that scores provide a quantitative evaluation of quality, while feedback offers a qualitative assessment.Both aspects reflect essay quality from different perspectives, and they are inherently consistent and can reinforce each other.In this paper, we propose a novel bidirectional reinforcement learning framework that effectively utilizes this consistency constraint to jointly optimize score prediction and feedback generation, ensuring mutual reinforcement and alignment between them.In this way, our model is hope to obtain a simultaneous accurate ratings and consistent text feedback.We conducted extensive experiments on publicly available datasets.The results demonstrate that our approach surpasses the current state-of-theart models, enhancing both scoring accuracy and feedback quality.

Health information fact-checking has been identified as a potential solution to combat the spread of misinformation on social media. Therefore, it is crucial to understand what factors are associated with individuals’ fact-checking behaviors on health (mis)information. This study examined how two types of social media involvement (active vs. passive) affected individuals’ fact-checking behaviors. An online survey was conducted to collect the data and test the hypotheses. The results indicated that passive and active social media involvement were negatively and positively associated with health information fact-checking, respectively. Further analysis revealed that the association between the two ways of social media involvement and health information fact-checking was mediated by social media fatigue and moderated by self-perceived social media information literacy. These findings can help us to understand the complex effects of social media involvement on fact-checking behavior, and provide guidance to design campaigns aimed at promoting health information fact-checking.