6 papers
Synthetic Vasculature and Pathology Enhance Vision-Language Model Reasoning
Chenjun Li, Cheng Wan, Laurin Lux +4
Vision-Language Models (VLMs) offer a promising path toward interpretable medical diagnosis by allowing users to ask about clinical explanations alongside predictions and across di…
PRISM-Bench: A Benchmark of Puzzle-Based Visual Tasks with CoT Error Detection
Yusu Qian, Cheng Wan, Chao Jia +3
Multimodal large language models (MLLMs) have achieved remarkable progress on vision-language tasks, yet their reasoning processes remain sometimes unreliable. We introduce PRISM-B…
Review of Inference-Time Scaling Strategies: Reasoning, Search and RAG
Zhichao Wang, Cheng Wan, Dong Nie
The performance gains of LLMs have historically been driven by scaling up model size and training data. However, the rapidly diminishing availability of high-quality training data…
veScale: Consistent and Efficient Tensor Programming with Eager-Mode SPMD
Youjie Li, Cheng Wan, Zhiqi Lin +10
Large Language Models (LLMs) have scaled rapidly in size and complexity, requiring increasingly intricate parallelism for distributed training, such as 3D parallelism. This sophist…
WASABI: A Metric for Evaluating Morphometric Plausibility of Synthetic Brain MRIs
Bahram Jafrasteh, Wei Peng, Cheng Wan +3
Generative models enhance neuroimaging through data augmentation, quality improvement, and rare condition studies. Despite advances in realistic synthetic MRIs, evaluations focus o…
MixGCN: Scalable GCN Training by Mixture of Parallelism and Mixture of Accelerators
Cheng Wan, Runkai Tao, Zheng Du +2
Graph convolutional networks (GCNs) have demonstrated superiority in graph-based learning tasks. However, training GCNs on full graphs is particularly challenging, due to the follo…