Guangxuan Xiao

37 papers A* 8Journal 25Unranked 4

Year	Rank	Type	Title / Venue / Authors
2025	J	`jnl`	Accelerating Large-Scale Reasoning Model Inference with Sparse Self-Speculative Decoding. CoRR Yilong Zhao, Jiaming Tang, Kan Zhu, Zihao Ye, Chi-Chih Chang, Chaofan Lin, Jongseok Park, Guangxuan Xiao, Mohamed S. Abdelfattah, Mingyu Gao, Baris Kasikci, Song Han, Ion Stoica
2025	J	`jnl`	BLASST: Dynamic BLocked Attention Sparsity via Softmax Thresholding. CoRR Jiayi Yuan, Cameron Shinn, Kai Xu, Jingze Cui, George Klimiashvili, Guangxuan Xiao, Perkz Zheng, Bo Li, Yuxin Zhou, Zhouhai Ye, Weijie You, Tian Zheng, Dominic Brown, Pengbo Wang, Richard Cai, Julien Demouth, John D. Owens, Xia Hu, Song Han, Timmy Liu, Huizi Mao
2025	A*	`conf`	DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads. ICLR Guangxuan Xiao, Jiaming Tang, Jingwei Zuo, Junxian Guo, Shang Yang, Haotian Tang, Yao Fu, Song Han
2025	J	`jnl`	FastComposer: Tuning-Free Multi-subject Image Generation with Localized Attention. Int. J. Comput. Vis. Guangxuan Xiao, Tianwei Yin, William T. Freeman, Frédo Durand, Song Han
2025	J	`jnl`	Four Over Six: More Accurate NVFP4 Quantization with Adaptive Block Scaling. CoRR Jack Cook, Junxian Guo, Guangxuan Xiao, Yujun Lin, Song Han
2025	—	`conf`	LServe: Efficient Long-sequence LLM Serving with Unified Sparse Attention. MLSys Shang Yang, Junxian Guo, Haotian Tang, Qinghao Hu, Guangxuan Xiao, Jiaming Tang, Yujun Lin, Zhijian Liu, Yao Lu, Song Han
2025	J	`jnl`	LServe: Efficient Long-sequence LLM Serving with Unified Sparse Attention. CoRR Shang Yang, Junxian Guo, Haotian Tang, Qinghao Hu, Guangxuan Xiao, Jiaming Tang, Yujun Lin, Zhijian Liu, Yao Lu, Song Han
2025	J	`jnl`	Optimizing Mixture of Block Attention. CoRR Guangxuan Xiao, Junxian Guo, Kasra Mazaheri, Song Han
2025	—	`conf`	QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving. MLSys Yujun Lin, Haotian Tang, Shang Yang, Zhekai Zhang, Guangxuan Xiao, Chuang Gan, Song Han
2025	A*	`conf`	Retrieval Head Mechanistically Explains Long-Context Factuality. ICLR Wenhao Wu, Yizhong Wang, Guangxuan Xiao, Hao Peng, Yao Fu
2025	J	`jnl`	StreamingVLM: Real-Time Understanding for Infinite Video Streams. CoRR Ruyi Xu, Guangxuan Xiao, Yukang Chen, Liuning He, Kelly Peng, Yao Lu, Song Han
2025	A*	`conf`	XAttention: Block Sparse Attention with Antidiagonal Scoring. ICML Ruyi Xu, Guangxuan Xiao, Haofeng Huang, Junxian Guo, Song Han
2025	J	`jnl`	XAttention: Block Sparse Attention with Antidiagonal Scoring. CoRR Ruyi Xu, Guangxuan Xiao, Haofeng Huang, Junxian Guo, Song Han
2024	—	`conf`	AWQ: Activation-aware Weight Quantization for On-Device LLM Compression and Acceleration. MLSys Ji Lin, Jiaming Tang, Haotian Tang, Shang Yang, Wei-Ming Chen, Wei-Chen Wang, Guangxuan Xiao, Xingyu Dang, Chuang Gan, Song Han
2024	J	`jnl`	AWQ: Activation-aware Weight Quantization for On-Device LLM Compression and Acceleration. GetMobile Mob. Comput. Commun. Ji Lin, Jiaming Tang, Haotian Tang, Shang Yang, Guangxuan Xiao, Song Han
2024	A*	`conf`	BitDelta: Your Fine-Tune May Only Be Worth One Bit. NeurIPS James Liu, Guangxuan Xiao, Kai Li, Jason D. Lee, Song Han, Tri Dao, Tianle Cai
2024	J	`jnl`	BitDelta: Your Fine-Tune May Only Be Worth One Bit. CoRR James Liu, Guangxuan Xiao, Kai Li, Jason D. Lee, Song Han, Tri Dao, Tianle Cai
2024	J	`jnl`	Correction to: Red Alarm for Pre-trained Models: Universal Vulnerability to Neuron-level Backdoor Attacks. Mach. Intell. Res. Zhengyan Zhang, Guangxuan Xiao, Yongwei Li, Tian Lv, Fanchao Qi, Zhiyuan Liu, Yasheng Wang, Xin Jiang, Maosong Sun
2024	J	`jnl`	DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads. CoRR Guangxuan Xiao, Jiaming Tang, Jingwei Zuo, Junxian Guo, Shang Yang, Haotian Tang, Yao Fu, Song Han
2024	A*	`conf`	Efficient Streaming Language Models with Attention Sinks. ICLR Guangxuan Xiao, Yuandong Tian, Beidi Chen, Song Han, Mike Lewis
2024	J	`jnl`	FreshGNN: Reducing Memory Access via Stable Historical Embeddings for Graph Neural Network Training. Proc. VLDB Endow. Kezhao Huang, Haitian Jiang, Minjie Wang, Guangxuan Xiao, David Wipf, Xiang Song, Quan Gan, Zengfeng Huang, Jidong Zhai, Zheng Zhang
2024	A*	`conf`	InfLLM: Training-Free Long-Context Extrapolation for LLMs with an Efficient Context Memory. NeurIPS Chaojun Xiao, Pengle Zhang, Xu Han, Guangxuan Xiao, Yankai Lin, Zhengyan Zhang, Zhiyuan Liu, Maosong Sun
2024	J	`jnl`	InfLLM: Unveiling the Intrinsic Capacity of LLMs for Understanding Extremely Long Sequences with Training-Free Memory. CoRR Chaojun Xiao, Pengle Zhang, Xu Han, Guangxuan Xiao, Yankai Lin, Zhengyan Zhang, Zhiyuan Liu, Song Han, Maosong Sun
2024	J	`jnl`	QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving. CoRR Yujun Lin, Haotian Tang, Shang Yang, Zhekai Zhang, Guangxuan Xiao, Chuang Gan, Song Han
2024	A*	`conf`	QUEST: Query-Aware Sparsity for Efficient Long-Context LLM Inference. ICML Jiaming Tang, Yilong Zhao, Kan Zhu, Guangxuan Xiao, Baris Kasikci, Song Han
2024	J	`jnl`	Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference. CoRR Jiaming Tang, Yilong Zhao, Kan Zhu, Guangxuan Xiao, Baris Kasikci, Song Han
2024	J	`jnl`	Retrieval Head Mechanistically Explains Long-Context Factuality. CoRR Wenhao Wu, Yizhong Wang, Guangxuan Xiao, Hao Peng, Yao Fu
2023	J	`jnl`	Efficient Streaming Language Models with Attention Sinks. CoRR Guangxuan Xiao, Yuandong Tian, Beidi Chen, Song Han, Mike Lewis
2023	J	`jnl`	FastComposer: Tuning-Free Multi-Subject Image Generation with Localized Attention. CoRR Guangxuan Xiao, Tianwei Yin, William T. Freeman, Frédo Durand, Song Han
2023	J	`jnl`	Offsite-Tuning: Transfer Learning without Full Model. CoRR Guangxuan Xiao, Ji Lin, Song Han
2023	J	`jnl`	ReFresh: Reducing Memory Access from Exploiting Stable Historical Embeddings for Graph Neural Network Training. CoRR Kezhao Huang, Haitian Jiang, Minjie Wang, Guangxuan Xiao, David Wipf, Xiang Song, Quan Gan, Zengfeng Huang, Jidong Zhai, Zheng Zhang
2023	J	`jnl`	Red Alarm for Pre-trained Models: Universal Vulnerability to Neuron-level Backdoor Attacks. Mach. Intell. Res. Zhengyan Zhang, Guangxuan Xiao, Yongwei Li, Tian Lv, Fanchao Qi, Zhiyuan Liu, Yasheng Wang, Xin Jiang, Maosong Sun
2023	A*	`conf`	SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models. ICML Guangxuan Xiao, Ji Lin, Mickaël Seznec, Hao Wu, Julien Demouth, Song Han
2023	J	`jnl`	Sparse and Local Networks for Hypergraph Reasoning. CoRR Guangxuan Xiao, Leslie Pack Kaelbling, Jiajun Wu, Jiayuan Mao
2022	J	`jnl`	SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models. CoRR Guangxuan Xiao, Ji Lin, Mickaël Seznec, Julien Demouth, Song Han
2022	—	`conf`	Sparse and Local Networks for Hypergraph Reasoning. LoG Guangxuan Xiao, Leslie Pack Kaelbling, Jiajun Wu, Jiayuan Mao
2021	J	`jnl`	Red Alarm for Pre-trained Models: Universal Vulnerabilities by Neuron-Level Backdoor Attacks. CoRR Zhengyan Zhang, Guangxuan Xiao, Yongwei Li, Tian Lv, Fanchao Qi, Zhiyuan Liu, Yasheng Wang, Xin Jiang, Maosong Sun