In this work, we examine hateful memes from three complementary angles - how to detect them, how to explain their content and how to intervene them before being posted - by applying a range of strategies built on top of generative AI models. To the best of our knowledge, explanation and intervention have typically been studied separately from detection, which does not reflect real-world conditions. Further, since curating large annotated datasets for meme moderation is prohibitively expensive, we propose a novel framework - PEST - that leverages task-specific generative VLMs and the few-shot adaptability of large VLMs to cater to different types of memes. We believe this is the first work focused on generalizable hateful meme moderation under limited data conditions, and has strong potential for deployment in real-world production scenarios. Warning: Contains potentially toxic contents.
Data-centric scaling of RLVR has focused on collecting or generating verifiable training problems. However, corpus size alone does not determine whether the dataset provides a useful learning signal for a target policy. To guide the synthesis of effective RLVR data, we experimentally characterize dataset effectiveness...
Ximing Lu, Aviral Kumar, Jan Kautz et al.· 0 citations
Multimodal models have often scored below text-only baselines in the BabyLM Challenge (Klerings et al., 2024; Takmaz et al., 2025). We test whether additional recurrent computation helps close this gap. Recurrence increases computation by reusing layers without adding parameters or training data. Using a single impleme...
Daichi Hotta, Yohei Kobashi, Takeshi Kojima et al.· Proceedings of the Second Ba...· 0 citations
Interleaved multimodal in-context learning (ICL) lets frozen large vision-language models (LVLMs) acquire new tasks from a handful of image-text-answer demonstrations, yet it remains fragile: demonstrations bind to the wrong images, redundant exemplars crowd out informative ones, and noisy answers are copied verbatim i...
Y. Yılmaz, Elena Petrova, M. Kaya et al.· 0 citations
While Large Language Models (LLMs) possess rich world knowledge and impressive generalization capabilities, their direct application to tabular data classification is hindered by high inference costs and limited interpretability. In contrast, decision trees are fast and transparent but often underperform in low-data re...
Yue Qiu, Ze-Kang Du, Yi-Qun Diao et al.· Proceedings of the 2026 Conf...· 0 citations
In-context learning (ICL) enables language models to perform new tasks from demonstrations without weight updates. However, every ICL inference requires processing the full set of examples, resulting in inefficient deployments, and how ICL works mechanistically is not fully understood. Prior work compresses ICL into fi...
Guang-Zhi Xiong, Zheng-Hao He, Bo-Han Liu et al.· 0 citations
In generative recommendation, position has two distinct meanings: an event’s order in the sequence and its wall-clock time. Existing approaches usually inject temporal information through auxiliary embeddings or relative attention biases, while vanilla RoPE captures only order. We revisit RoPE for generative recommenda...
Xiao-Kai Wei, Jia-Jun Wu, Dai-Yao Yi et al.· Proceedings of the 20th ACM...· 2 citations· ⚡1
The new ChartNet training dataset could improve the accuracy of vision-language models that help analyze business trends or interpret scientific figures.
