Jingcheng Hu

representation learning (1) vision transformer (1) transfer learning (1) mathematical reasoning (1) self-supervised learning (1) attention mechanism (1) model pretraining (1) low-rank approximation (1) message passing (1) kv cache (1) masked image modeling (1) fine-tuning performance (1) test-time compute (1) multi-matrix factorization attention (1) parallel coordinated reasoning (1) outcome-based reinforcement learning (1)