Search papers, labs, and topics across Lattice.
Shanghai Jiao Tong University
2
0
3
E^3RL not only overcomes the autoregressive curse but also enhances LLMs' reasoning capabilities, achieving up to 6.514% better performance than previous state-of-the-art models.
LLM multi-agent systems can achieve significantly higher accuracy at a fraction of the cost by learning to selectively delegate tasks instead of relying on rigid orchestration.
