Train What You Deploy:Token-Faithful Post-Training of a Production Coding
This work presents a fidelity-aware training coupling framework that retains trainer-side sampling over original prompts, eliminates spurious model calls via a negotiated training protocol, and restricts loss computation to verifiable token spans with closed-failure guarantees.
Cheng Li, Jie-Xi Liu, Yi-Xuan Chen et al.
· 0 citations