Search papers, labs, and topics across Lattice.
Kuaishou Technology
1
0
3
Aligning LLM reasoning with a dedicated recommendation head via reinforcement learning yields state-of-the-art recommendation performance in real-world systems.
