Xingjian Diao, Chunhui Zhang, Keyi Kong, Weiyi Wu, Chiyu Ma, Zhongyu Ouyang, Peijun Qing, Soroush Vosoughi, Jiang Gui. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

Automated detection of intake gestures with wearable sensors has been a critical area of research for advancing our understanding and ability to intervene in people's eating behavior. Numerous algorithms have been developed and evaluated in terms of accuracy. However, ensuring the system is not only accurate in making predictions but also efficient in doing so is critical for real-world deployment. Despite the growing research on accurate detection of intake gestures using wearables, many of these algorithms are often energy inefficient, impeding on-device deployment for continuous and real-time monitoring of diet. This paper presents a template-based optimized multicenter classifier that enables accurate intake gesture detection while maintaining low-inference time and energy consumption using a wrist-worn accelerometer and gyroscope. We designed an Intake Gesture Counter smartphone application (CountING) and validated the practicality of our algorithm against seven state-of-the-art approaches on three public datasets (In-lab FIC, Clemson, and OREBA). Compared with other methods, we achieved optimal accuracy (81.60% F1 score) and very low inference time (15.97 msec per 2.20-sec data sample) on the Clemson dataset, and among the top performing algorithms, we achieve comparable accuracy (83.0% F1 score compared with 85.6% in the top performing algorithm) but superior inference time (13.8x faster, 33.14 msec per 2.20-sec data sample) on the In-lab FIC dataset and comparable accuracy (83.40% F1 score compared with 88.10% in the top-performing algorithm) but superior inference time (33.9x faster, 16.71 msec inference time per 2.20-sec data sample) on the OREBA dataset. On average, our approach achieved a 25-hour battery lifetime (44% to 52% improvement over state-of-the-art approaches) when tested on a commercial smartwatch for continuous real-time detection. Our approach demonstrates an effective and efficient method, enabling real-time intake gesture detection using wrist-worn devices in longitudinal studies.

Xingjian Diao, Tianzhen Yang, Chunhui Zhang, Weiyi Wu, Ming Cheng, Jiang Gui. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers). 2025.

As large language models (LLMs) are evolving at a rapid pace, the design space for building high-performance and efficient models has expanded significantly. This position paper argues that the next level of LLMs will combine decoder-only architectures with modular enhancements in perception and specialization, supported by adaptive post-training and hardware-aware optimization strategies. We analyze 31 influential LLMs developed by 12 leading institutions, examining their design choices across three critical dimensions: model architecture, post-training, and optimization. By asking and answering 9 fundamental research questions surrounding these design choices, we identify key trends and propose forward-looking directions to guide the design of next-generation LLMs that are more capable, efficient, and adaptable to diverse real-world needs. To support the community, we maintain a GitHub repository that will be continuously updated at https://github.com/meettyj/LLMs-Design-Choices .