The field of social media analysis is rapidly evolving, with a focus on detecting and combating misinformation. Recent research has emphasized the importance of stance detection, with innovative methods being proposed to enhance the accuracy of stance prediction. These methods often incorporate emotional awareness, label fusion, and dual cross-attention mechanisms to capture the nuances of social media content. Additionally, there is a growing interest in multimodal rumor detection, which aims to align intrinsic and social modalities to improve the detection of rumors. The impact of COVID-19 on Twitter ego networks has also been studied, revealing temporary changes in network structure, sentiment, and topics during lockdown periods. Furthermore, research has explored the role of conditional reasoning in answer set programming and the identification of linguistic fingerprints of online users who engage with conspiracy communities. Noteworthy papers in this area include: ISMAF, which proposes a novel framework for multimodal rumor detection, CL-ISR, which combines contrastive learning and implicit stance reasoning for misleading text detection, SPLAENet, which achieves state-of-the-art results in stance detection using a dual cross-attentive neural network with label fusion.