Related papers: Dupo: A Mixed-Initiative Authoring Tool for Respon…
We present LongVPO, a novel two-stage Direct Preference Optimization framework that enables short-context vision-language models to robustly understand ultra-long videos without any long-video annotations. In Stage 1, we synthesize…
Many recommendation systems limit user inputs to text strings or behavior signals such as clicks and purchases, and system outputs to a list of products sorted by relevance. With the advent of generative AI, users have come to expect richer…
While current emotional support dialogue systems typically rely on expert-defined scalar rewards for alignment, these signals suffer from severe information sparsity. They cannot explain why a response failed or how to adapt to dynamic user…
Embodied agents have achieved prominent performance in following human instructions to complete tasks. However, the potential of providing instructions informed by texts and images to assist humans in completing tasks remains underexplored.…
Navigation in new or unknown environments is vital, especially for visually impaired individuals. While many solutions exist, few are tailored to specific disabilities, often due to limited collaboration with handicap users in the design…
Preference optimization is a critical post-training technique used to align large language models (LLMs) with human preferences, typically by fine-tuning on ranked response pairs. While methods like Direct Preference Optimization (DPO) have…
Individuals with Intellectual and Developmental Disabilities (IDD) have unique needs and challenges when working with data. While visualization aims to make data more accessible to a broad audience, our understanding of how to design…
Interactive information dashboards can help both specialists and the general public understand complex datasets; but interacting with these dashboards often presents users with challenges such as understanding and verifying the presented…
The design of a building requires an architect to balance a wide range of constraints: aesthetic, geometric, usability, lighting, safety, etc. At the same time, there are often a multiplicity of diverse designs that can meet these…
Most software applications contain graphics such as charts, diagrams and maps. Currently, these graphics are designed with a ``one size fits all" approach and do not cater to the needs of people with disabilities. Therefore, when using…
Visual language reasoning requires a system to extract text or numbers from information-dense images like charts or plots and perform logical or arithmetic reasoning to arrive at an answer. To tackle this task, existing work relies on…
Findings from graphical perception can guide visualization recommendation algorithms in identifying effective visualization designs. However, existing algorithms use knowledge from, at best, a few studies, limiting our understanding of how…
Multi-agent workflows have become an effective strategy for tackling complicated tasks by decomposing them into multiple sub-tasks and assigning them to specialized agents. However, designing optimal workflows remains challenging due to the…
Composite visualization represents a widely embraced design that combines multiple visual representations to create an integrated view. However, the traditional approach of creating composite visualizations in immersive environments…
Information visualization significantly enhances human perception by graphically representing complex data sets. The variety of visualization designs makes it challenging to efficiently evaluate all possible designs catering to users'…
Pre-trained vision-language models like CLIP have remarkably adapted to various downstream tasks. Nonetheless, their performance heavily depends on the specificity of the input text prompts, which requires skillful prompt template…
Multi-turn human-AI collaboration is fundamental to deploying interactive services such as adaptive tutoring, conversational recommendation, and professional consultation. However, optimizing these interactions via reinforcement learning is…
Characterizing creativity in visualization design can lead to the design of more expressive representations and visualization authoring tools that prioritize human creativity. In this paper, we examine how creativity manifests itself in…
Optimization is fundamental across numerous disciplines, typically following an iterative process of refining an initial solution to enhance performance. This principle is equally critical in prompt engineering, where designing effective…
The muti-modal or multi-sensorial perception of nature is presented in this article as part of research devoted to inclusive tools developed in the framework of User Centered Design. This proposal shows that it is possible to work in a…