vix.ing
·
top
·
new
·
best
·
stats
·
spec
Anhao Zhao
Post-Training Shifts Confidence: A Three-Stage Analysis of How SFT, RL, and OPD Shape CoT Calibration
2026/07/15 by
Shuhao Li
,
Guodong Du
,
Anhao Zhao
+3
#cs.CL