5 papers
MMCIG: Multimodal Cover Image Generation for Text-only Documents and Its Dataset Construction via Pseudo-labeling
Hyeyeon Kim, Sungwoo Han, Jingun Kwon +2
In this study, we introduce a novel cover image generation task that produces both a concise summary and a visually corresponding image from a given text-only document. Because no…
Length Representations in Large Language Models
Sangjun Moon, Dasom Choi, Jingun Kwon +2
Large language models (LLMs) have shown remarkable capabilities across various tasks, that are learned from massive amounts of text-based data. Although LLMs can control output seq…
gMBA: Expression Semantic Guided Mixed Boolean-Arithmetic Deobfuscation Using Transformer Architectures
Youjeong Noh, Joon-Young Paik, Jingun Kwon +1
Mixed Boolean-Arithmetic (MBA) obfuscation protects intellectual property by converting programs into forms that are more complex to analyze. However, MBA has been increasingly exp…
Diversity of Transformer Layers: One Aspect of Parameter Scaling Laws
Hidetaka Kamigaito, Ying Zhang, Jingun Kwon +3
Transformers deliver outstanding performance across a wide range of tasks and are now a dominant backbone architecture for large language models (LLMs). Their task-solving performa…
Considering Length Diversity in Retrieval-Augmented Summarization
Juseon-Do, Jaesung Hwang, Jingun Kwon +2
This study investigates retrieval-augmented summarization by specifically examining the impact of exemplar summary lengths under length constraints, not covered by previous work. W…