8 papers
When Sinks Help or Hurt: Unified Framework for Attention Sink in Large Vision-Language Models
Jiho Choi, Jaemin Kim, Sanghwan Kim +2
Attention sinks are defined as tokens that attract disproportionate attention. While these have been studied in single modality transformers, their cross-modal impact in Large Visi…
PosterForest: Hierarchical Multi-Agent Collaboration for Scientific Poster Generation
Jiho Choi, Seojeong Park, Seongjong Song +1
Automating scientific poster generation requires hierarchical document understanding and coherent content-layout planning. Existing methods often rely on flat summarization or opti…
Mitigating Perceptual Judgment Bias in Multimodal LLM-as-a-Judge via Perceptual Perturbation and Reward Modeling
Seojeong Park, Jiho Choi, Junyong Kang +3
Recent multimodal large language models have demonstrated strong reasoning ability, yet their reliability as automated evaluators remains limited by a critical weakness: when visua…
What "Not" to Detect: Negation-Aware VLMs via Structured Reasoning and Token Merging
Inha Kang, Youngsun Lim, Seonho Lee +3
State-of-the-art vision-language models (VLMs) suffer from a critical failure in understanding negation, often referred to as affirmative bias. This limitation is particularly seve…
MomentMix Augmentation with Length-Aware DETR for Temporally Robust Moment Retrieval
Seojeong Park, Jiho Choi, Kyungjune Baek +1
Video Moment Retrieval (MR) aims to localize moments within a video based on a given natural language query. Given the prevalent use of platforms like YouTube for information retri…
3D-Aware Vision-Language Models Fine-Tuning with Geometric Distillation
Seonho Lee, Jiho Choi, Inha Kang +3
Vision-Language Models (VLMs) have shown remarkable performance on diverse visual and linguistic tasks, yet they remain fundamentally limited in their understanding of 3D spatial s…