Assistant Professor, KAIST
5 papers at NeurIPS 2025
We propose Flex-Judge, a multimodal judge model trained solely on a small corpus of high-quality text reasoning data.
We propose KLASS, a fast KL-guided sampling method for masked diffusion models that improves accuracy while cutting inference time by over 2x.
AdaSTaR enhances STaR by using adaptive sampling for diversity and curriculum to reduce training data imbalance, achieving best accuracy across six benchmarks while reducing training FLOPs by 58.6%.
