Asymmetric Temperature Scaling Makes Larger Networks Teach Well Again Xin-Chun Li1 Wen-Shu Fan1 Shaoming Song2 Yinchuan Li2
AsymmetricTemperatureScalingMakesLargerNetworksTeachWellAgainXin-ChunLi1,Wen-ShuFan1,ShaomingSong2,YinchuanLi2BingshuaiLi2,YunfengShao2,De-ChuanZhan11StateKeyLaboratoryforNovelSoftwareTechnology,NanjingUniversity,Nanjing,China2HuaweiNoah'sArkLab,Beijing,China{lixc,fanws}@lamda.nju.edu.cn,zhandc@nju....
2025-05-06
1.42MB 21 页 0
0
10玖币