Haoran Xie

Vision-Language Pre-Training (VLP) aims to convert vision-language tasks into image-text matching challenges, necessitating robust interaction between visual and textual modalities. However, current research often faces a tradeoff: methods pursuing fine-grained alignment typically rely on computationally expensive object detectors and pre-defined annotations, while coarse-grained alignment methods overlook subtle visual details. To alleviate these issues, we present OMEGA, a C O mprehensive M ulti-l E vel G ranularity A lignment framework. OMEGA achieves sufficient cross-modal alignment without relying on external object detectors or annotations. Specifically, we first design a Text-Aware Patch Selection (TAPS) module, which dynamically constructs patch-level regions for fine-grained excavation based on informative token selection, effectively bypassing the need for bounding boxes. Second, to facilitate multi-grained information fusion, we introduce a Cross-Grained Alignment (CGA) module to learn modality-shared features across global and object levels. Experimental results demonstrate that OMEGA surpasses existing approaches in five pivotal vision-language tasks—image-text retrieval, visual question answering, visual reasoning, visual entailment, and image captioning, while maintaining superior inference efficiency compared to detector-based methods. We believe this study offers a promising object-free paradigm for scalable vision-language pre-training.

Retrieval-based dialogue systems aim to select a proper response according to multi-turn conversational history. Persona-based conversation utilizes prior knowledge to maintain persona consistency, enhancing retrieval accuracy. However, reference-based evaluation relies on high-quality data annotation, which is costly and time-consuming. To address this, we discover persona-centric metamorphic relations to infer test samples from annotated data, without additional annotation cost. Benefiting from this, this work efficiently evaluates the robustness of personalized dialogue models regarding persona consistency. Specifically, we discover three types of metamorphic relations from three aspects: self-persona, partner-persona, and response, to automatically derive new test samples . Then the inherent inference relations between originals and derivatives allow for robustness evaluation. Using this evaluation methodology, our work assesses three widely used training paradigms: non-pretraining, fine-tuning after pre-training, and prompt learning, in personalized dialogue retrieval to observe whether these paradigms are more robust or exhibit the same flaws as the other two paradigms. Our experimental results, based on the three discovered metamorphic relations with consistent outputs reveal that prompt learning is more robust than training from scratch and fine-tuning. While traditional reference-based validation and natural language processing methods achieve competitively high retrieval accuracy (Hits@1 up to 87.4%), the persona consistency of dialogue retrieval systems is just 20.98% when persona descriptions are perturbed using various metamorphic relation-based transformations.

Artificial intelligence in online learning (AIOL) has attracted increasing attention in academia. This study applied BERTopic, a transformer-based topic modeling approach that leverages contextual embeddings, to examine the thematic structure and evolution of AIOL research. By analyzing 1,048 AIOL publications, this study sought to answer three research questions. What are the major research topics in AIOL? How have these topics evolved over time? What future research directions are recognized? Several research themes were identified, such as adaptive learning systems, sentiment analysis, and predictive analytics. Temporal analysis revealed a shift from early applications of traditional AI toward machine learning and deep learning approaches. The results also revealed an increasing emphasis on multimodal data integration for emotion recognition and personalized learning support. Based on the findings, a conceptual model has been proposed to guide AIOL research and practice by integrating four components: data, AI processing, adaptive learning, and learner development. Overall, the study offered a data-driven overview of AIOL research and demonstrated how topic modeling can be used to examine thematic development in emerging research fields.