Abstract:While pre-trained language models excel at semantic understanding, they often struggle to capture nuanced affective information critical for affective recognition tasks. To address these limitations, we propose a novel framework for enhancing emotion-aware embeddings in transformer-based models. Our approach introduces a continuous valence-arousal labeling system to guide contrastive learning, which captures subtle and multi-dimensional emotional nuances more effectively. Furthermore, we employ a dynamic token perturbation mechanism, using gradient-based saliency to focus on sentiment-relevant tokens, improving model sensitivity to emotional cues. The experimental results demonstrate that the proposed framework outperforms existing methods, achieving up to 15.5% improvement in the emotion classification benchmark, highlighting the importance of employing continuous labels. This improvement demonstrates that the proposed framework is effective in affective representation learning and enables precise and contextually relevant emotional understanding.
Abstract:Manual optimization of traffic light cycles is a complex and time-consuming task, necessitating the development of automated solutions. In this paper, we propose the application of reinforcement learning to optimize traffic light cycles in real-time. We present a case study using the Simulation Urban Mobility simulator to train a Deep Q-Network algorithm. The experimental results showed 44.16% decrease in the average number of Emergency stops, showing the potential of our approach to reduce traffic congestion and improve traffic flow. Furthermore, we discuss avenues for future research and enhancements to the reinforcement learning model.