Researcher — paper.dinq.me

2026 顶会论文

作者位次 #7/11ICML 2026Poster

LoKiFormer: Locality-aware Attention with Decoupled Knowledge Memory for Efficient Large Language Model Pretraining

Qiuwu Chen · Zimo Liu · Yuchen Li · Ying Sun · Yifan Zhang · Zhijie Qiu · Zeng You · Ryan Dong · Simeng Ma · Yaofo Chen · Mingkui Tan

检索与知识效率与系统数据与预训练OpenReview

一作ACL 2026Poster· Natural Language Processing

Latent-Condensed Transformer for Efficient Long Context Modeling

Zeng You · Yaofo Chen · Qiuwu Chen · Ying Sun · Shuhai Zhang · Yingjian Li · Yaowei Wang · Mingkui Tan

摘要

Large language models (LLMs) face significant challenges in processing long contexts due to the linear growth of the key-value (KV) cache and quadratic complexity of self-attention. Existing approaches address these bottlenecks separately: Multi-head Latent Attention (MLA) reduces the KV cache by...

效率与系统长上下文与架构可解释性OpenReview 项目页

作者位次 #4/7AAAI 2026Poster· Artificial Intelligence

Action-and-object Aware Alignment for Partially Relevant Video Retrieval

Chuanshen Chen · Kai Zhou · Zhiquan Wen · Zeng You · Yirui Li · Tianhang Xiang · Mingkui Tan

对齐与偏好多模态检索与知识OpenReview

作者位次 #3/4AAAI 2026Poster· Artificial Intelligence

ProCache: Constraint-Aware Feature Caching with Selective Computation for Diffusion Transformer Acceleration

Fanpu Cao · Yaofo Chen · Zeng You · Wei Luo

OpenReview

Zeng You

获取完整研究档案

联系方式

2026 顶会论文

LoKiFormer: Locality-aware Attention with Decoupled Knowledge Memory for Efficient Large Language Model Pretraining

Latent-Condensed Transformer for Efficient Long Context Modeling

Action-and-object Aware Alignment for Partially Relevant Video Retrieval

ProCache: Constraint-Aware Feature Caching with Selective Computation for Diffusion Transformer Acceleration

常见合作者