collaborators

8 papers

cs.LG2026

Demystifying Transition Matching: When and Why It Can Beat Flow Matching

Jaihoon Kim, Rajarshi Saha, Minhyuk Sung +1

Flow Matching (FM) underpins many state-of-the-art generative models, yet recent results indicate that Transition Matching (TM) can achieve higher quality with fewer sampling steps…

cs.LG2026

Scaling Laws Meet Model Architecture: Toward Inference-Efficient LLMs

Song Bian, Tao Yu, Shivaram Venkataraman +1

Scaling the number of parameters and the size of training data has proven to be an effective strategy for improving large language model (LLM) performance. Yet, as these models gro…

cs.LG2026

Online Posterior Sampling with a Diffusion Prior

Branislav Kveton, Boris Oreshkin, Youngsuk Park +2

Posterior sampling in contextual bandits with a Gaussian prior can be implemented exactly or approximately using the Laplace approximation. The Gaussian prior is computationally ef…

cs.LG2025

MuonBP: Faster Muon via Block-Periodic Orthogonalization

Ahmed Khaled, Kaan Ozkara, Tao Yu +2

Gradient orthogonalization is a simple strategy that shows great utility in speeding up gradient descent. The Muon optimizer (Jordan, Jin, et al., 2024) combines gradient orthogona…

cs.LG2025

Training LLMs with MXFP4

Albert Tseng, Tao Yu, Youngsuk Park

Low precision (LP) datatypes such as MXFP4 can accelerate matrix multiplications (GEMMs) and reduce training costs. However, directly using MXFP4 instead of BF16 during training si…

cs.LG2025

RoSTE: An Efficient Quantization-Aware Supervised Fine-Tuning Approach for Large Language Models

Quan Wei, Chung-Yiu Yau, Hoi-To Wai +4

Supervised fine-tuning is a standard method for adapting pre-trained large language models (LLMs) to downstream tasks. Quantization has been recently studied as a post-training tec…