Multimodal metaphor detection is an important and challenging task in multimedia computing, which aims to distinguish between metaphorical and literal multimodal expressions. Existing studies mainly utilize typical multimodal computing approaches for detection, neglecting the unique cross-domain and cross-modality characteristics underlying multimodal metaphor understanding. According to Conceptual Metaphor Theory (CMT), the inconsistency between source and target domains and their attribute similarity are essential to infer the intricate meanings implied in metaphors. In practice, the scarcity of the annotated multimodal metaphorical contents in the real world brings additional difficulty to the detection task and further complicates the understanding of multimodal metaphors. To address the above challenges, in this paper, we propose a novel Imaginative FRame Augmented (ImaRA) method for low-resource multimodal metaphor detection and explanation inspired by CMT. Specifically, we first identify imaginative frame as an associative structure to stimulate the imaginative thinking of multimodal metaphor detection and understanding. We then construct a cross-modal imagination dataset rich in multimodal metaphors and corresponding imaginative frames, and retrieve an augmented instance from this imagination dataset using imaginative frames mined from the input. This augmented instance serves as the demonstration exemplar to boost the metaphor reasoning ability of the multimodal large language model (MLLM) in low-resource multimodal scenarios. Experiments on two publicly available datasets show that our method consistently achieves robust results compared to MLLM-based methods for both multimodal metaphor detection and explanation in low-resource scenarios and meanwhile surpasses existing multimodal metaphor detection methods with full training data.

Stance detection aims to identify the attitudes toward specific targets from text, which is an important research area in text mining and social media analytics. Existing research is mainly conducted in monolingual setting on English datasets. To tackle the data scarcity problem in low-resource languages, cross-lingual stance detection (CLSD) transfers the knowledge from high-resource (source) language to low-resource (target) language. The CLSD task is the most challenging in zero-shot setting when no training data is available in target language, and transferring stance-relevant knowledge learned from high-resource language to bridge the language gap is the key for improving the performance of zero-shot CLSD. In this paper, we leverage the capability of large language model (LLM) for stance knowledge acquisition, and propose KEAR, a knowledge elicitation and retrieval framework. The knowledge elicitation module in KEAR first derives different types of stance knowledge from LLM’s reasoning process. Then, the knowledge retrieval module in KEAR matches the target language input to the most relevant stance knowledge for enhancing text representations. Experiments on multilingual datasets show the effectiveness of KEAR compared with competitive baselines as well as the CLSD approaches trained with labeled data in target language.

Multimodal sarcasm detection is an important research topic in natural language processing and multimedia computing, and benefits a wide range of applications in multiple domains. Most existing studies regard the incongruity between image and text as the indicative clue in identifying multimodal sarcasm. To capture cross-modal incongruity, previous methods rely on fixed architectures in network design, which restricts the model from dynamically adjusting to diverse image-text pairs. Inspired by routing-based dynamic network, we model the dynamic mechanism in multimodal sarcasm detection and propose the Dynamic Routing Transformer Network (DynRT-Net). Our method utilizes dynamic paths to activate different routing transformer modules with hierarchical co-attention adapting to cross-modal incongruity. Experimental results on a public dataset demonstrate the effectiveness of our method compared to the state-of-the-art methods. Our codes are available at https://github.com/TIAN-viola/DynRT.

Stance detection is an important task in text mining and social media analytics, aiming to automatically identify the user’s attitude toward a specific target from text, and has wide applications in a variety of domains. Previous work on stance detection has mainly focused on monolingual setting. To address the problem of imbalanced language resources, cross-lingual stance detection is proposed to transfer the knowledge learned from a high-resource (source) language (typically English) to another low-resource (target) language. However, existing research on cross-lingual stance detection has ignored the inconsistency in the occurrences and distributions of targets between languages, which consequently degrades the performance of stance detection in low-resource languages. In this paper, we first identify the target inconsistency issue in cross-lingual stance detection, and propose a fine-grained Target-oriented Relation Alignment (TaRA) method for the task, which considers both target-level associations and language-level alignments. Specifically, we propose the Target Relation Graph to learn the in-language and cross-language target associations. We further devise the relation alignment strategy to enable knowledge transfer between semantically correlated targets across languages. Experimental results on the representative datasets demonstrate the effectiveness of our method compared to competitive methods under variant settings.