On-policy self-distillation (OPSD) combines student generated trajectories with dense supervision from a privileged teacher of the same model. Its pointwise clipping protects the student from locally incompatible teacher directions, but the fixed teacher receives no feedback about the supervision the student rejects, w...
Under shared-history workloads, Cached-Linear substantially reduces computation and memory compared with Global-Softmax, while Balanced GDN improves selected direct-prediction endpoints with efficient context reuse.
Guang-Fu Guo, Xiao-Qian Lu, Rui Liu et al.· 0 citations
Real-time decision-making for enhanced geothermal systems (EGS) is challenging because long-term production periods involve high-dimensional control spaces and a large number of time-consuming high-fidelity hydrothermal simulations. Reinforcement learning provides a natural framework for state-dependent sequential cont...
Ruimin Dai, Guodong Chen, Randy Harsuko et al.· 0 citations
DualOPSD, an asymmetric alternating framework that adapts both policies, reduces truncation across all three scales, and makes later supervision responsive to the learner and does not require another rollout.
DeltaFlow is a promising alternative to dense attention for efficient continuous language denoising and noise-adaptive memory control and scheduled Temporal State Consistency to stabilize hidden representations across nearby noise levels are introduced.
Guang-Fu Guo, Xiao-Qian Lu, Linsey Pang et al.· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
