Deploying Large Language Models (LLMs) directly on mobile platforms at the edge is gaining a lot of traction due to a myriad of benefits, such as increased privacy, personalization, and reduced latency. However, LLMs have heavy computational requirements, which are difficult for resource-constrained mobile and edge platforms to fulfill. In addition to limited compute resources, mobile and edge systems often have a compact form factor and lack physical mechanisms to dissipate heat generated from high processor usage rates (e.g., fans) to prevent throttling and reduced processing power, which LLMs can easily cause. To mitigate these effects, prior works have proposed various power governing strategies, such as dynamic voltage and frequency scaling (DVFS), for reducing power and heat generation for heavy computational tasks on mobile platforms. Recently, DVFS methods tailored for mobile LLMs have also been proposed. However, these methods mostly focus on optimizing hardware parameters and processor frequencies, and they fall short under some thermally constrained scenarios. Drawing from recent advances in machine learning, we identify and take advantage of the key insight that not all tokens require full-depth inference to maintain high-quality generation. Motivated by this, we present PELM, a solution that augments traditional DVFS processor frequency tuning with two additional workload-specific knobs: 1) speculative decoding and 2) variable verification depth to expand the optimization space to multiple dimensions for more power efficient on-device LLM inference. In extensive evaluations across different hardware platforms and datasets, PELM demonstrates superior performance compared to state-of-the-art power governing methods, with up to 23.1% speedup and 52.4% reduction in energy consumption, while maintaining comparable task performance. The source code is available at https://github.com/imec-nu/PELM.

The existing EEG emotion classification methods have some problems, such as insufficient emotion representation and lack of targeted channel enhancement module due to feature redundancy. To this end, a novel EEG emotion recognition method (SCLE-2D-CNN) combining scaled convolutional layer (SCLs), enhanced channel module and two-dimensional convolutional neural network (2D-CNN) is proposed. Firstly, the time-frequency features of multi-channel EEG emotional signals were extracted by stacking scl layer by layer. Secondly, channel enhancement module is used to reassign different importance to all EEG physical channels. Finally, 2D-CNN was used to obtain deep local spatiotemporal features and complete emotion classification. The experimental results show that the accuracy of SEED data set and F1 are 98.09% and 97.00%, respectively, and the binary classification accuracy of DEAP data set is 98.06% and 96.83%, respectively, which are superior to other comparison methods. The proposed method has a certain application prospect in the recognition of human mental state.