From łog π to π: Taming Divergence in Soft Clipping via Bilateral Decoupled Decay of Probability Gradient Weight.
Xiaoliang Fu, Jiaye Lin, Yangyi Fang, Chaowen Hu, Cong Qin, Zekai Shao, Binbin Zheng, Lu Pan, Ke Zeng
Browse the full ACL paper archive.