Zhengyan Zhang

发表论文 48 篇 · 总被引 10096 次 · h-index 21

代表论文

• DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence (2026 · 被引 518)
• InfLLM: Training-Free Long-Context Extrapolation for LLMs with an Efficient Context Memory (2024 · Neural Information Processing Systems · 被引 187)
• ReLU2 Wins: Discovering Efficient Activation Functions for Sparse LLMs (2024 · arXiv.org · 被引 76)
• Finding Skill Neurons in Pre-trained Transformer-based Language Models (2022 · Conference on Empirical Methods in Natural Language Processing · 被引 76)
• InfLLM: Unveiling the Intrinsic Capacity of LLMs for Understanding Extremely Long Sequences with Training-Free Memory (2024 · arXiv.org · 被引 74)
• ProSparse: Introducing and Enhancing Intrinsic Activation Sparsity within Large Language Models (2024 · International Conference on Computational Linguistics · 被引 53)