SearchWiki: Learning to Build and Navigate Knowledge Wikis for Active Information Seeking
SearchWiki paired with WikiResearcher-9B demonstrates that learned navigation over structured corpora is a superior alternative to flat retrieval, and optimizing the agent's navigation policy with on-policy reinforcement learning with a multi-component reward function balancing answer correctness, retrieval quality and...
Guransh Singh, Vishwajeet Kumar, Arkadeep Acharya et al.
· 0 citations