Context-attended Adversarial Reinforcement Learning for Robust Multi-step Retrieval Augmented Generation.

Yingtao Ren, Xiao Luo, Yu-Cheng Chang, Chin-Teng Lin

Browse the full ACL paper archive.