Yiming Zhang

5 papers Nanyang Technological University

• InfiFPO: Implicit Model Fusion via Preference Optimization in Large Language Models
• InfiGFusion: Graph-on-Logits Distillation via Efficient Gromov-Wasserstein for Model Fusion
• Multi-Scale Finetuning for Encoder-based Time Series Foundation Models
• Table as a Modality for Large Language Models
• Towards Large-Scale In-Context Reinforcement Learning by Meta-Training in Randomized Worlds