ReNeg
ReNeg copied to clipboard
ReNeg: Learning Negative Embedding with Reward Guidance