Search papers, labs, and topics across Lattice.
School of Data Science, The Chinese University of Hong Kong, Shenzhen, China, LIGHTSPEED
2
0
2
2
MRRG reveals that leveraging multiple evaluative perspectives can significantly enhance the quality of reward signals for LLM optimization, outperforming traditional single-role approaches.
Deep clustering's reign is not yet here: conventional methods still dominate, especially when paired with pretrained embeddings.
