Search papers, labs, and topics across Lattice.
School of Computer Science and Technology, Tianjin University, Tianjin, China
2
0
5
Targeted neuron fine-tuning can unlock superior image translation capabilities in multimodal large language models, outperforming traditional methods by preserving pre-trained knowledge.
LLM serving can achieve 5.6x higher throughput without sacrificing latency by decoupling preemption granularity from scheduling frequency.
