Tianjin Huang

Dual-Kernel Adapter: Expanding Spatial Horizons for Data-Constrained Medical Image Analysis

Efficient and Stable Grouped RL Training for Large Language Models

LOST: Low-rank and Sparse Pre-training for Large Language Models

Stable-SPAM: How to Stably Train Large Language Models in 4-Bit

Compositional Architecture of Regret in Large Language Models

Enhancing Robust Fairness via Confusional Spectral Regularization

LIFT the Veil for the Truth: Principal Weights Emerge after Rank Reduction for Reasoning-Focused Supervised Fine-Tuning

SPAM: Spike-Aware Adam with Momentum Reset for Stable LLM Training

Composable Interventions for Language Models