Shengnan Wang
3 papers · 2025–2026 · 2 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓
π
Conference Polyglot
(2)
π
Interdisciplinary Bridge
πΊοΈ
Taxonomy Completionist
(10)
π§
Keyword Pioneer
π
Cross-Pollinator
(15)
Conferences
AAAI (2)
ACL (1)
Top co-authors
Keywords
model inference
(2)
mixture of expert
(2)
dynamic routing
(1)
neural network optimization
(1)
efficient computing
(1)
binary hashing
(1)
communication efficiency
(1)
pipeline parallelism
(1)
inference optimization
(1)
sparse attention
(1)
inference acceleration
(1)
expert parallelism
(1)
large language model
(1)
learn to hash
(1)
model compression
(1)
top-k attention
(1)
distributed learning
(1)
Papers
SMIDT: High-Performance Inference Framework for MoE Models with Dynamic Top-K Routing
AAAI 2026
BigMac: A Communication-Efficient Mixture-of-Experts Model Structure for Fast Training and Inference
AAAI 2025
HATA: Trainable and Hardware-Efficient Hash-Aware Top-k Attention for Scalable Large Model Inference
ACL 2025