5 papers
Accelerating Prefilling via Decoding-time Contribution Sparsity
Zhiyuan He, Yike Zhang, Chengruidong Zhang +3
Large Language Models (LLMs) incur quadratic attention complexity with input length, creating a major time bottleneck in the prefilling stage. Existing acceleration methods largely…
Efficient Attention Mechanisms for Large Language Models: A Survey
Yutao Sun, Zhenyu Li, Yike Zhang +4
Transformer-based architectures have become the prevailing backbone of large language models. However, the quadratic time and memory complexity of self-attention remains a fundamen…
VL Norm: Rethink Loss Aggregation in RLVR
Zhiyuan He, Xufang Luo, Yike Zhang +2
We propose VL Norm (Variance-reduced Length-dependent Normalization), a simple yet effective loss aggregation method tailored to the characteristic of dynamic generation lengths in…
LeanK: Learnable K Cache Channel Pruning for Efficient Decoding
Yike Zhang, Zhiyuan He, Huiqiang Jiang +4
Large language models (LLMs) enable long-context tasks but face efficiency challenges due to the growing key-value (KV) cache. We propose LeanK, a learning-based method that prunes…
FocusLLM: Precise Understanding of Long Context by Dynamic Condensing
Zhenyu Li, Yike Zhang, Tengyu Pan +6
Empowering LLMs with the ability to precisely understand long contexts is crucial for many downstream applications. However, handling long contexts with conventional transformer ar…