7 papers
When Irrelevant Text Matters: Affine Margin Shifts in Multimodal Large Language Models
Yinfeng Wang, Zhiyuan Yao, Zheren Fu +2
Multimodal large language models (MLLMs) are frequently exposed to auxiliary textual context, the impact of which on visually grounded tasks remains underexplored. In this paper, w…
Uncertainty-Aware Exploratory Direct Preference Optimization for Multimodal Large Language Models
Huatian Zhang, Zhendong Mao, Lei Zhang +1
Direct Preference Optimization (DPO) has proven to be an effective solution for mitigating hallucination in Multimodal Large Language Models (MLLMs) by learning from preference pai…
CreatiParser: Generative Image Parsing of Raster Graphic Designs into Editable Layers
Weidong Chen, Dexiang Hong, Zhendong Mao +4
Graphic design images consist of multiple editable layers, such as text, background, and decorative elements, while most generative models produce rasterized outputs without explic…
FACE-net: Factual Calibration and Emotion Augmentation for Retrieval-enhanced Emotional Video Captioning
Weidong Chen, Cheng Ye, Zhendong Mao +5
Emotional Video Captioning (EVC) is an emerging task, which aims to describe factual content with the intrinsic emotions expressed in videos. Existing works perceive global emotion…
LayerEdit: Disentangled Multi-Object Editing via Conflict-Aware Multi-Layer Learning
Fengyi Fu, Mengqi Huang, Lei Zhang +1
Text-driven multi-object image editing which aims to precisely modify multiple objects within an image based on text descriptions, has recently attracted considerable interest. Exi…
DiT: Dynamic Diffusion Transformer for Accurate Image Generation
Weinan Jia, Mengqi Huang, Nan Chen +2
Diffusion models are widely recognized for their ability to generate high-fidelity images. Despite the excellent performance and scalability of the Diffusion Transformer (DiT) arch…
CustomContrast: A Multilevel Contrastive Perspective For Subject-Driven Text-to-Image Customization
Nan Chen, Mengqi Huang, Zhuowei Chen +3
Subject-driven text-to-image (T2I) customization has drawn significant interest in academia and industry. This task enables pre-trained models to generate novel images based on uni…