7 papers
From Seeing to Thinking: Decoupling Perception and Reasoning Improves Post-Training of Vision-Language Models
Juncheng Wu, Hardy Chen, Haoqin Tu +6
Recent advances in vision-language models (VLMs) emphasize long chain-of-thought reasoning; yet, we find that their performance on visual tasks is primarily limited by a lack of vi…
Real Images, Worse Judgments: Evaluating Vision-Language Models on Concreteness and Imagery
Yifan Jiang, Ruoxi Ning, Sheng Yao +1
Visual inputs are often assumed to improve language understanding in multimodal models. We examine this assumption by asking whether vision-language models (VLMs) can distinguish u…
Chartographer: Counterfactual Chart Generation for Evaluating Vision-Language Models
Yifan Jiang, Dae Yon Hwang, Jesse C. Cresswell +1
Chart question-answering (QA) benchmarks aim to pose questions that require visual reasoning to correctly answer, but vision-language models (VLMs) can often reach solutions throug…
From Tokens to Numbers: Continuous Number Modeling for SVG Generation
Michael Ogezi, Martin Bell, Freda Shi +1
For certain image generation tasks, vector graphics such as Scalable Vector Graphics (SVGs) offer clear benefits such as increased flexibility, size efficiency, and editing ease, b…
A Very Big Video Reasoning Suite
Maijunxian Wang, Ruisi Wang, Juyi Lin +53
Rapid progress in video models has largely focused on visual quality, leaving their reasoning capabilities underexplored. Video reasoning grounds intelligence in spatiotemporally c…
SpaRE: Enhancing Spatial Reasoning in Vision-Language Models with Synthetic Data
Michael Ogezi, Freda Shi
Vision-language models (VLMs) work well in tasks ranging from image captioning to visual question answering (VQA), yet they struggle with spatial reasoning, a key skill for underst…
Logical forms complement probability in understanding language model (and human) performance
Yixuan Wang, Freda Shi
With the increasing interest in using large language models (LLMs) for planning in natural language, understanding their behaviors becomes an important research question. This work…