Zhenpeng Su
8 papers · 2024–2026 · 4 venues · across top CS/AI venues
Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.
Achievements
Jump to papers ↓
🌍
Conference Polyglot · Bronze
🌉
Interdisciplinary Bridge · Bronze
🗺️
Taxonomy Completionist
(26)
⚡
Prolific Year · Productive
Top co-authors
Research topics
Machine Learning
(7)
Artificial Intelligence
(5)
Natural Language Processing
(5)
Core AI
(5)
Deep Learning
(4)
Application Areas
(4)
Optimization & Theory · Machine Learning
(4)
Generation
(3)
Optimization & Theory · Deep Learning
(3)
Applications
(2)
Learning Types · Machine Learning
(2)
Resources & Methods
(2)
Architectures
(1)
Learning Types · Deep Learning
(1)
Models
(1)
Keywords
language model
(2)
large language model
(2)
model compression
(2)
language modeling
(2)
mixture of expert
(2)
entropy regularization
(1)
information retrieval
(1)
hyperparameter optimization
(1)
exploration-exploitation tradeoff
(1)
dense retrieval
(1)
information entropy
(1)
distributionally robust optimization
(1)
knowledge distillation
(1)
loss function
(1)
semantic alignment
(1)
scaling law
(1)
dynamic routing
(1)
sparse activation
(1)
proximal policy optimization
(1)
mathematical reasoning
(1)
Papers
CE-GPPO: Coordinating Entropy via Gradient-Preserving Clipping Policy Optimization in Reinforcement Learning
ACL 2026
CartesianMoE: Boosting Knowledge Sharing among Experts via Cartesian Product Routing in Mixture-of-Experts
NAACL 2025
DSMoE: Matrix-Partitioned Experts with Dynamic Routing for Computation-Efficient Dense LLMs
EMNLP 2025
Scaffold-BPE: Enhancing Byte Pair Encoding for Large Language Models with Simple and Effective Scaffold Token Removal
AAAI 2025
Task-level Distributionally Robust Optimization for Large Language Model-based Dense Retrieval
AAAI 2025
Temporal Scaling Law for Large Language Models
EMNLP 2025
Dial-MAE: ConTextual Masked Auto-Encoder for Retrieval-based Dialogue Systems
NAACL 2024
MiLe Loss: a New Loss for Mitigating the Bias of Learning Difficulties in Generative Language Models
NAACL 2024