AdaTutoRank: Learning to Rerank Document Sets via Adaptive Tutoring Optimization for RAG and Deep Research
AdaTutoRank is proposed, a setwise reranker trained with Adaptive Tutoring Optimization under a three-level hierarchy of nine rubric dimensions, which supplies silver labels for the cold start, rewards for reinforcement learning, and hints for distillation.
Kai-Lin Jiang, Lei Liu, Jian-Fei Xi et al.
· 0 citations