Can Thinking Models Think to Detect Hateful Memes?

Hateful memes often require compositional multimodal reasoning: the image and text may appear benign in isolation, yet their interaction conveys harmful intent. Although thinking-based multimodal large language models (MLLMs) have recently advanced vision–language understanding, their capabilities remain underexplored for hateful meme analysis. We propose a reinforcement learning–based post-training framework that improves reasoning in thinking-based MLLMs via task-specific rewards and a novel Group Relative Policy Optimization (GRPO) objective. Concretely, we (i) conduct a systematic empirical study of off-the-shelf MLLMs for hateful meme understanding, (ii) extend an existing hateful meme dataset by generating weakly/pseudo-supervised chain-of-thought (CoT) rationales via distillation, and (iii) introduce a GRPO-based objective that jointly optimizes meme classification and explanation quality to encourage fine-grained step-by-step reasoning. Experiments on the Hateful Memes benchmark show that our approach achieves state-of-the-art results, improving accuracy and F1 by approximately 1% and explanation quality by approximately 3%. We will publicly release our code, data extensions, and evaluation resources to support reproducibility.