Search papers, labs, and topics across Lattice.
School of Computer Science and Technology, Tianjin University 鈭桬qual contribution.
2
0
4
SharQ recovers up to 63% of the accuracy gap in low-bit quantization while achieving over 2x latency reduction in LLM inference.
ID-LoRA slashes trainable parameters by up to 46% compared to standard LoRA while boosting performance across diverse benchmarks, offering a sweet spot between efficiency and effectiveness for fine-tuning LLMs.
