SPPO: Sequence-Level PPO for Long-Horizon Reasoning Tasks
Paper • 2604.08865 • Published • 29
TIANYI
BIMU233
AI & ML interests
None yet
Recent Activity
upvoted a paper about 16 hours ago
Conditional Equivalence of DPO and RLHF: Implicit Assumption, Failure Modes, and Provable Alignment upvoted a paper 14 days ago
Continuous Latent Diffusion Language ModelOrganizations
None yet
authored a
paper about 1 month ago