9 papers · 1 filter
Beyond Atomic Layouts: Compositional Design Understanding with Vision-Language Models
Yiyang Huang, Zhaowen Wang, Simon Jenni +4
Layout understanding, or the interpretation of element organization, is essential for document analysis, user interface (UI) creation, and graphic design. While recent vision-langu…
Hierarchical Visual Agent: Managing Contexts in Joint Image-Text Space for Advanced Chart Reasoning
Qihua Dong, Ruozhen He, Junwen Chen +4
Advanced chart question answering requires both precise perception of small visual elements and multi-step reasoning across several subplots. While existing MLLMs are strong at und…
UnAC: Adaptive Visual Prompting with Abstraction and Stepwise Checking for Complex Multimodal Reasoning
Yifan Wang, Yun Fu
Although recent LMMs have become much stronger at visual perception, they remain unreliable on problems that require multi-step reasoning over visual evidence. In this paper, we pr…
Ref-Adv: Exploring MLLM Visual Reasoning in Referring Expression Tasks
Qihua Dong, Kuo Yang, Lin Ju +6
Referring Expression Comprehension (REC) links language to region level visual perception. Standard benchmarks (RefCOCO, RefCOCO+, RefCOCOg) have progressed rapidly with multimodal…
Seeing Through Words: Controlling Visual Retrieval Quality with Language Models
Jianglin Lu, Simon Jenni, Kushal Kafle +3
Text-to-image retrieval is a fundamental task in vision-language learning, yet in real-world scenarios it is often challenged by short and underspecified user queries. Such queries…
Diffusion-DRF: Free, Rich, and Differentiable Reward for Video Diffusion Fine-Tuning
Yifan Wang, Yanyu Li, Gordon Guocheng Qian +3
Video diffusion alignment has been heavily relied on scalar rewards. These rewards are typically derived from learned reward models in human preference datasets, requiring addition…
Consistency and Uncertainty: Identifying Unreliable Responses From Black-Box Vision-Language Models for Selective Visual Question Answering
Zaid Khan, Yun Fu
The goal of selective prediction is to allow an a model to abstain when it may not be able to deliver a reliable prediction, which is important in safety-critical contexts. Existin…
Self-Training Large Language Models for Improved Visual Program Synthesis With Visual Reinforcement
Zaid Khan, Vijay Kumar BG, Samuel Schulter +2
Visual program synthesis is a promising approach to exploit the reasoning abilities of large language models for compositional computer vision tasks. Previous work has used few-sho…
Exploring Question Decomposition for Zero-Shot VQA
Zaid Khan, Vijay Kumar BG, Samuel Schulter +2
Visual question answering (VQA) has traditionally been treated as a single-step task where each question receives the same amount of effort, unlike natural human question-answering…