Speech emotion recognition (SER), a rapidly evolving task that aims to recognize the emotion of speakers, has become a key research area in affective computing. However, various languages in multilingual natural scenarios extremely challenge the generalization ability of SER, causing the model performance to decrease quickly, and driving researchers to ask how to improve the performance of multilingual SER. Recent studies mainly use feature fusion and language-controlled models to address this challenge, but key points such as the intrinsic association of languages or deep analysis of multilingual shared features (MSFs) are still neglected. To solve this problem, an explainable Multitask-based Shared Feature Learning (MSFL) model is proposed for multilingual SER. The introduction of multi-task learning (MTL) can provide related task information of language recognition for MSFL, improve its generalization in multilingual situations, and further lay the foundation for learning MSFs. Specifically, considering the generalization capability and interpretability of the model, the powerful MTL module was combined with the long short-term memory and attention mechanism, aiming to maintain the generalization in multilingual situations. Then, the feature weights acquired from the attention mechanism were ranked in descending order, and the top-ranked MSFs were compared with top-ranked monolingual features, enhancing the model interpretability based on the feature comparison. Various experiments were conducted on Emo-DB, CASIA, and SAVEE corpora from the model generalization and interpretability aspects. Experimental results indicate that MSFL performs better than most state-of-the-art models, with an average improvement of 3.37–4.49%. Besides, the top 10 features in MSFs almost contain the top-ranked features in three monolingual features, which effectively demonstrates the interpretability of MSFL.
With the rapid development of Multimodal Large Language Models (MLLMs) in education, their applications have mainly focused on content generation tasks such as text writing and courseware production. However, automated assessment of non-exam learning outcomes remains underexplored. This study shifts the application of MLLMs from content generation to content evaluation and designs a lightweight and extensible framework to enable automated assessment of students’ multimodal work. We constructed a multimodal dataset comprising student essays, slide decks, and presentation videos from university students, which were annotated by experts across five educational dimensions. Based on horizontal educational evaluation dimensions (Format Compliance, Content Quality, Slide Design, Verbal Expression, and Nonverbal Performance) and vertical model capability dimensions (consistency, stability, and interpretability), we systematically evaluated four leading multimodal large models (GPT-4o, Gemini 2.5, Doubao1.6, and Kimi 1.5) in assessing non-exam learning outcomes. The results indicate that MLLMs demonstrate good consistency with human evaluations across various assessment dimensions, with each model exhibiting its own strengths. Additionally, they possess high explainability and perform better in text-based tasks than in visual tasks, but their scoring stability still requires improvement. This study demonstrates the potential of MLLMs for non-exam learning assessment and provides a reference for advancing their applications in education.
In this paper, we present the first systematic investigation and quantification of Similar Object Interference (SOI), a long-overlooked yet critical bottleneck in Single Object Tracking (SOT). Through controlled Online Interference Masking (OIM) experiments, we quantitatively demonstrate that eliminating interference sources leads to substantial performance improvements (AUC gains up to 4.35) across all SOTA trackers, directly validating SOI as a primary constraint for robust tracking and highlighting the feasibility of external cognitive guidance. Building upon these insights, we adopt natural language as a practical form of external guidance, and construct SOIBench-the first semantic cognitive guidance benchmark specifically targeting SOI challenges. It automatically mines SOI frames through multi-tracker collective judgment and introduces a multi-level annotation protocol to generate precise semantic guidance texts. Systematic evaluation on SOIBench reveals a striking finding: existing vision-language tracking (VLT) methods fail to effectively exploit semantic cognitive guidance, achieving only marginal improvements or even performance degradation (AUC changes of -0.26 to +0.71). In contrast, we propose a novel paradigm employing large-scale vision-language models (VLM) as external cognitive engines that can be seamlessly integrated into arbitrary RGB trackers. This approach demonstrates substantial improvements under semantic cognitive guidance (AUC gains up to 0.93), representing a significant advancement over existing VLT methods. We hope SOIBench will serve as a standardized evaluation platform to advance semantic cognitive tracking research and contribute new insights to the tracking research community.
20 August 2026 in International Journal of Educational Research