List your company

Conversational Bandit on PET-inspired conversational bandit downstream tasks

95RewardTask-only
48.260.3572.584.65May 20, 2026

Evaluation results

MethodLinks
Task-only2026.05950.24
RCWeight ω=0.252026.05940.22
Ours2026.05750
RCWeight ω=0.752026.05500