Skip to content

Author

Yi-Hang Chen

We have 2 of 6 papers

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

ReSPO: Reshaped Sequence Policy Optimization for Gradient Starvation in Off-Policy Learning

This work proposes ReSPO (Reshaped Sequence Policy Optimization), a smooth, two-branch sequence-level kernel derived from an $\alpha$-divergence variational objective and an exponential variance-control tilt that effectively learns from long positive reasoning trajectories during early training, even when accumulated p...

Yi-Hang Chen, Yuan-Hao Ban, Cho-Jui Hsieh · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.