101 citations · 235 across the 20 of their papers we have counts for
14 papers · 1 filter
SparDA: Sparse Decoupled Attention for Efficient Long-Context LLM Inference
Yaosheng Fu, Guangxuan Xiao, Xin Dong +2
Sparse attention reduces compute and memory bandwidth for long-context LLM inference. However, two key challenges remain: (1) KV cache capacity still grows with sequence length, an…
BLASST: Dynamic BLocked Attention Sparsity via Softmax Thresholding
Jiayi Yuan, Cameron Shinn, Kai Xu +19
The growing demand for long-context inference capabilities in Large Language Models (LLMs) has intensified the computational and memory bottlenecks inherent to the self-attention m…
Four Over Six: More Accurate NVFP4 Quantization with Adaptive Block Scaling
Jack Cook, Junxian Guo, Guangxuan Xiao +7
As large language models have grown larger, interest has grown in low-precision numerical formats such as NVFP4 as a way to improve speed and reduce memory usage. However, quantizi…
XAttention: Block Sparse Attention with Antidiagonal Scoring
Ruyi Xu, Guangxuan Xiao, Haofeng Huang +2
Long-Context Transformer Models (LCTMs) are vital for real-world applications but suffer high computational costs due to attention's quadratic complexity. Block-sparse attention mi…
LServe: Efficient Long-sequence LLM Serving with Unified Sparse Attention
Shang Yang, Junxian Guo, Haotian Tang +7
Large language models (LLMs) have shown remarkable potential in processing long sequences and complex reasoning tasks, yet efficiently serving these models remains challenging due…
DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads
Guangxuan Xiao, Jiaming Tang, Jingwei Zuo +5
Deploying long-context large language models (LLMs) is essential but poses significant computational and memory challenges. Caching all Key and Value (KV) states across all attenti…
Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference
Jiaming Tang, Yilong Zhao, Kan Zhu +3
As the demand for long-context large language models (LLMs) increases, models with context windows of up to 128K or 1M tokens are becoming increasingly prevalent. However, long-con…
QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving
Yujun Lin, Haotian Tang, Shang Yang +4
Quantization can accelerate large language model (LLM) inference. Going beyond INT8 quantization, the research community is actively exploring even lower precision, such as INT4. N…
Retrieval Head Mechanistically Explains Long-Context Factuality
Wenhao Wu, Yizhong Wang, Guangxuan Xiao +2
Despite the recent progress in long-context language models, it remains elusive how transformer-based models exhibit the capability to retrieve relevant information from arbitrary…
InfLLM: Training-Free Long-Context Extrapolation for LLMs with an Efficient Context Memory
Chaojun Xiao, Pengle Zhang, Xu Han +5
Large language models (LLMs) have emerged as a cornerstone in real-world applications with lengthy streaming inputs (e.g., LLM-driven agents). However, existing LLMs, pre-trained o…
Efficient Streaming Language Models with Attention Sinks
Guangxuan Xiao, Yuandong Tian, Beidi Chen +2
Deploying Large Language Models (LLMs) in streaming applications such as multi-round dialogue, where long interactions are expected, is urgently needed but poses two major challeng…
Offsite-Tuning: Transfer Learning without Full Model
Guangxuan Xiao, Ji Lin, Song Han
Transfer learning is important for foundation models to adapt to downstream tasks. However, many foundation models are proprietary, so users must share their data with model owners…
SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models
Guangxuan Xiao, Ji Lin, Mickael Seznec +3
Large language models (LLMs) show excellent performance but are compute- and memory-intensive. Quantization can reduce memory and accelerate inference. However, existing methods ca…
Red Alarm for Pre-trained Models: Universal Vulnerability to Neuron-Level Backdoor Attacks
Zhengyan Zhang, Guangxuan Xiao, Yongwei Li +6
Pre-trained models (PTMs) have been widely used in various downstream tasks. The parameters of PTMs are distributed on the Internet and may suffer backdoor attacks. In this work, w…