This paper introduces Causal Chain Generation (CCG), a novel task where models have to generate the sequence of intermediate events connecting a cause and an effect, and fine-tune VLMs with GRPO using a composite reward that combines a format reward, SODA-Embedding, and Causal-QA accuracy.
R. Ghermi, Vicky Kalogeiton, Ivan Laptev· Proceedings of the 2026 Conf...· 0 citations
This work proposes Short-Films 20K (SF20K), the largest publicly available movie dataset, and accompanies this dataset with SF20K-Test, a manual, open-ended question answering benchmark, showing that instruction tuning on the large-scale dataset substantially improves model performance.
Ridouane Ghermi, Xi Wang, Vicky Kalogeiton et al.· International Journal of Com...· 11 citations· ⚡1
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.