Get started
Author

Zhenxu Tian

Observed name: Tian, Zhenxu
Observed institutions: Soochow University, Baidu (China)
OpenAlex ID: A5025685427
LongFlow: Efficient KV Cache Compression for Reasoning Models
2026 · Yi Su, Zhenxu Tian, et al. · Conference on Empirical Methods in Natural Language Processing (EMNLP) · 5
PDF
Where Matters More Than What: Decoding-aligned KV Cache Compression via Position-aware Pseudo Queries
2026 · Tian, Zhenxu, Su, Yi, et al. · arXiv (Cornell University) · 3
PDF
Prism Attention: Proposal-Refined Index Sharing Mechanism for Efficient LLMs Inference
2026 · Tian, Zhenxu, Liu, Kebin, et al. · arXiv (Cornell University)
PDF
Works count: 3
Citation count: 8
H-index: 2
I10-index:
Year
Topic
Type