Engaging in counterarguments (CAs) has been shown to be a promising approach to effectively support the development of critical thinking skills in educational settings. Developing an argumentation tutoring system that automatically provides personalized feedback on CAs would benefit both teachers and students by mitigating the constraints imposed by limited instructional resources. However, the development of such systems remains underexplored. To address this gap, we develop CAT, a CounterArgument Tutoring system that analyzes the underlying logical structure of CAs and provides tailored feedback accordingly. To evaluate CAT, we conduct a user study with 39 undergraduate students over two weeks in a real classroom, comparing it against an existing argumentation tutoring system as a baseline. The results indicate that CAT supports students in critically evaluating opposing arguments, considering alternative perspectives, and refining their own reasoning. Statistical analysis further shows that CAT elicits more engaged interaction than the baseline system. Moreover, students perceive CAT as easy to use and valuable for improving both their critical thinking and argumentation skills.

Seed science is essential for modern agriculture, directly influencing crop yields and global food security. However, challenges such as interdisciplinary complexity and high costs with limited returns hinder progress, leading to a shortage of experts and insufficient technological support. While large language models (LLMs) have shown promise across various fields, their application in seed science remains limited due to the scarcity of digital resources, complex gene-trait relationships, and the lack of standardized benchmarks. To address this gap, we introduce SeedBench—the first multi-task benchmark specifically designed for seed science. Developed in collaboration with domain experts, SeedBench focuses on seed breeding and simulates key aspects of modern breeding processes. We conduct a comprehensive evaluation of 26 leading LLMs, encompassing proprietary, open-source, and domain-specific fine-tuned models. Our findings not only highlight the substantial gaps between the power of LLMs and the real-world seed science problems, but also make a foundational step for research on LLMs for seed design.

Voice conversion has emerged as a pivotal technology in numerous applications ranging from assistive communication to entertainment. In this paper, we present RT-VC, a zero-shot real-time voice conversion system that delivers ultra-low latency and high-quality performance. Our approach leverages an articulatory feature space to naturally disentangle content and speaker characteristics, facilitating more robust and interpretable voice transformations. Additionally, the integration of differentiable digital signal processing (DDSP) enables efficient vocoding directly from articulatory features, significantly reducing conversion latency. Experimental evaluations demonstrate that, while maintaining synthesis quality comparable to the current state-of-the-art (SOTA) method, RT-VC achieves a CPU latency of 61.4 ms, representing a 13.3% reduction in latency.