Demo Data from EDBT 2026 / reviewers in the wild / expert
Zhenliang Xue
dblp:380/4390
· DBLP profile ↗
1ranked-venue papers
1as first author
1since 2021 · last 2026
0009-0004-4904-0078ORCID · reported
Domains — the database's venue-derived domains; a paper can count in several
Systems, architecture and hardware · 1 · 1 first-author · 1 since 2021Software engineering, systems software and programming languages · 1 · 1 first-author · 1 since 2021
Expertise — from the expertise taxonomy: the topics of the expert's papers under the CCF categories. A weight counts papers with recency: 1 for a paper about the topic, 0.3 when the topic is its context, halved every five years.
| Artificial intelligence
1 paper |
Efficient and distributed learning · 61% Deep learning architectures and training · 30% Vision and language · 9% |
Topics — the 4 heaviest of 4, each with the papers that count most for it
| Topic | Weight | Papers | Last | Evidence papers |
|---|---|---|---|---|
Machine learning › Efficient and distributed learning
distributed training |
1.0 | 1 | 2026 | DIP: Efficient Large Multimodal Model Training with Dynamic Interleaved Pipeline · ASPLOS (2) 2026 |
Machine learning › Deep learning architectures and training › multimodal deep learning
multimodal model training |
1.0 | 1 | 2026 | DIP: Efficient Large Multimodal Model Training with Dynamic Interleaved Pipeline · ASPLOS (2) 2026 |
Machine learning › Efficient and distributed learning › distributed training › model parallelism
pipeline parallelism |
1.0 | 1 | 2026 | DIP: Efficient Large Multimodal Model Training with Dynamic Interleaved Pipeline · ASPLOS (2) 2026 |
Computer vision › Vision and language › vision-language model
multimodal large language model |
0.3 | 1 | 2026 | DIP: Efficient Large Multimodal Model Training with Dynamic Interleaved Pipeline · ASPLOS (2) 2026 |
Methods — techniques the papers use, named apart from their topics
dynamic interleaved pipeline scheduling · 1.0
| Year | Publication | Venue | Position |
|---|---|---|---|
| 2026 | DIP: Efficient Large Multimodal Model Training with Dynamic Interleaved PipelineabstractLarge multimodal models (LMMs) have demonstrated excellent capabilities in both understanding and generation tasks with various modalities. While these models can accept flexible combinations of input data, their training efficiency suffers from two major issues: pipeline stage imbalance caused by heterogeneous model architectures, and training data dynamicity stemming from the diversity of multimodal data. Zhenliang Xue, Hanpeng Hu, Xing Chen 0009, Yixin Song 0003, Zeyu Mi, Yibo Zhu 0001, Daxin Jiang, Yubin Xia, Haibo Chen 0001 |
ASPLOS (2) | 1 |