Memes serve as a powerful medium of expression in the digital age, shaping cultural discourse and conveying ideas succinctly and engagingly. However, their potential for social abuse highlights the importance of developing effective methods to detect harmful content within memes. Recent studies on memes have focused on transforming images into textual captions using large language models (LLMs). However, these approaches often result in non-informative captions. Furthermore, previous methods have only been tested on limited datasets, providing insufficient evidence of their robustness. To address these limitations, we present a multimodal, agent-based framework designed to generate informative visual descriptions of memes by asking insightful questions to improve visual descriptions in zero-shot visual question-answering settings. Specifically, we leverage an LLM as agents with distinct roles and a large multimodal model (LMM) as a vision expert. These agents first analyze the images and then ask informative questions related to potential social abuse in memes to obtain high-quality answers about the images. Through continuous discussion guided by instructional prompts, the agents gather high-quality information while repeatedly acquiring image data from the LMM, which helps detect social abuse in memes. Results on a dataset of 6,626 memes across 5 tasks show our framework surpasses state-of-the-art methods, demonstrating strong generalizability and improved detection of social abuse in memes.

People are spending an enormous amount of time on dig-ital devices through graphical user interfaces (GUIs), e.g., computer or smartphone screens. Large language models (LLMs) such as ChatGPT can assist people in tasks like writing emails, but struggle to understand and interact with GUIs, thus limiting their potential to increase automation levels. In this paper, we introduce CogAgent, an 18-billion-parameter visual language model (VLM) specializing in GUI understanding and navigation. By utilizing both low-resolution and high-resolution image encoders, CogA-gent supports input at a resolution of1120 × 1120, enabling it to recognize tiny page elements and text. As a general-ist visual language model, CogAgent achieves the state of the art on five text-rich and four general VQA benchmarks, including VQAv2, OK- VQA, Text- Vqa, St- Vqa, ChartQA, infoVQA, DocVQA, MM-Vet, and POPE. CogAgent, using only screenshots as input, outperforms LLM-based methods that consume extracted HTML text on both PC and Android GUI navigation tasks-Mind2Web and AITW, ad-vancing the state of the art. The model and codes are available at https://github.com/THUDM/CogVLM.

We introduce CogVLM, a powerful open-source visual language foundation model. Different from the popular \emph{shallow alignment} method which maps image features into the input space of language model, CogVLM bridges the gap between the frozen pretrained language model and image encoder by a trainable visual expert module in the attention and FFN layers. As a result, CogVLM enables a deep fusion of vision language features without sacrificing any performance on NLP tasks. CogVLM-17B achieves state-of-the-art performance on 17 classic cross-modal benchmarks, including 1) image captioning datasets: NoCaps, Flicker30k, 2) VQA datasets: OKVQA, TextVQA, OCRVQA, ScienceQA, 3) LVLM benchmarks: MM-Vet, MMBench, SEED-Bench, LLaVABench, POPE, MMMU, MathVista, 4) visual grounding datasets: RefCOCO, RefCOCO+, RefCOCOg, Visual7W. Codes and checkpoints are available at Github.

Harmful memes detection is challenging due to the semantic gap between different modalities. Previous studies mainly focus on feature extraction and fusion to learn discriminative information from memes. However, they ignore the misalignment of the modalities caused by the modality gap and suffer from data scarcity, resulting in insufficient learning of fusion-based models. Recently, researchers transformed images into textual captions and used language models for predictions, resulting in non-informative image captions. To address these gaps, this paper proposes an instructions-based abstracting approach CapAlign, in zero-shot visual question-answering settings. Precisely, we prompt a large language model (LLM) to ask informative questions to a pre-trained vision-language model and use the dialogues to generate a high-quality image caption. Further, to align the generated caption with the textual content of a meme, we used an LLM with instructions to generate informative captions of the meme and then prepend it with the attributes of the visual content of a meme to a prompt-based LM for prediction. Experimental findings on two benchmark datasets show that our approach produces informative captions and outperforms state-of-the-art methods for detecting harmful memes.

The creative nature of memes has made it possible for harmful content to spread quickly and widely on the internet. Harmful memes can range from spreading hate speech promoting violence, and causing emotional distress to individuals or communities. These memes are often designed to be misleading, manipulative, and controversial, making it challenging to detect and remove them from online platforms. Previous studies focused on how to fuse visual and language modalities to capture contextual information. However, meme analysis still severely suffers from data deficiency, resulting in insufficient learning of fusion modules. Further, using conventional pretrained encoders for text and images exhibits a greater semantic gap in feature spaces and leads to low performance. To address these gaps, this paper reformulates a harmful meme analysis as an auto-filling and presents a prompt-based approach to identify harmful memes. Specifically, we first transform multimodal data to a single (i.e., textual) modality by generating the captions and attributes of the visual data and then prepend the textual data in the prompt-based pre-trained language model. Experimental results on two benchmark harmful memes datasets demonstrate that our method outperformed state-of-the-art methods. We conclude with the transferability and robustness of our approach to identify creative harmful memes.