Zangwei Zheng
13 papers · 2021–2025 · 9 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 ↓+5 more ↓ Show less ↑
π Interdisciplinary Bridge Β· Bronze πΊοΈ Taxonomy Completionist (29) π£ Hot Topic Early Bird π Conference Polyglot Β· Silver π€ Dynamic Duo (12)
π
Grand Slam
β‘
Prolific Year Β· Productive
β
The Questioner
π
Century Club Β· Bronze
ποΈ
Keyword Collector
(79)
Top co-authors
Research topics
Machine Learning
(11)
Artificial Intelligence
(8)
Core AI
(7)
Deep Learning
(6)
Application Areas
(5)
Optimization & Theory Β· Machine Learning
(5)
Learning Types Β· Machine Learning
(4)
Learning Types Β· Deep Learning
(3)
Natural Language Processing
(2)
Architectures
(2)
Optimization & Theory Β· Deep Learning
(2)
Resources & Methods
(2)
Computer Science
(1)
Core Methods
(1)
Learning Paradigms
(1)
Keywords
large batch training
(2)
catastrophic forgetting
(1)
sample selection
(1)
domain adaptation
(1)
self-supervised learning
(1)
in-context learning
(1)
cross-domain learning
(1)
language model training
(1)
multimodal learning
(1)
unsupervised domain adaptation
(1)
semantic alignment
(1)
neural network optimization
(1)
gradient estimation
(1)
max norm
(1)
few-shot learning
(1)
adaptive optimization
(1)
memory efficiency
(1)
model training
(1)
vision-language model
(1)
continual learning
(1)
Papers
DSP: Dynamic Sequence Parallelism for Multi-Dimensional Transformers
ICML 2025
MERIT: Maximum-normalized Element-wise Ratio for Language Model Large-batch Training
ICML 2025
Dataset Growth
ECCV 2024
How Does the Textual Information Affect the Retrieval of Multimodal In-Context Learning?
EMNLP 2024
InfoBatch: Lossless Training Speed Up by Unbiased Dynamic Data Pruning
ICLR 2024
OpenMoE: An Early Effort on Open Mixture-of-Experts Language Models
ICML 2024
A Study on Transformer Configuration and Training Objective
ICML 2023
CAME: Confidence-guided Adaptive Memory Efficient Optimization
ACL 2023
CowClip: Reducing CTR Prediction Model Training Time from 12 Hours to 10 Minutes on 1 GPU
AAAI 2023
Preventing Zero-Shot Transfer Degradation in Continual Learning of Vision-Language Models
ICCV 2023
Response Length Perception and Sequence Scheduling: An LLM-Empowered LLM Inference Pipeline
NeurIPS 2023
To Repeat or Not To Repeat: Insights from Scaling LLM under Token-Crisis
NeurIPS 2023
Prototypical Cross-Domain Self-Supervised Learning for Few-Shot Unsupervised Domain Adaptation
CVPR 2021