DiJia Su

SPG: Sandwiched Policy Gradient for Masked Diffusion Language Models

Why RL Updates Look Sparse: An Implicit Compass Drives Optimization Bias

Token Assorted: Mixing Latent and Text Tokens for Improved Language Model Reasoning

Training Large Language Models to Reason in a Continuous Latent Space

Training Large Language Model to Reason in a Continuous Latent Space

Dualformer: Controllable Fast and Slow Thinking by Learning with Randomized Reasoning Traces