Search papers, labs, and topics across Lattice.
2
0
5
0
DUPAR is proposed, a conversational retrieval framework with complementary slow and fast paths that significantly reduces retrieval errors when the previous turn retrieves correct evidence and the follow-up targets a one-hop neighboring chunk.
By weighting Q-learning updates based on action similarity, QSIM tames overestimation in multi-agent RL, leading to more stable and effective learning.
