collaborators

7 papers

cs.DC2026

Hetu v2: A General and Scalable Deep Learning System with Hierarchical and Heterogeneous Single Program Multiple Data Annotations

Haoyang Li, Fangcheng Fu, Hao Ge +6

The Single-Program Multiple-Data (SPMD) paradigm provides a unified abstraction to annotate various parallel dimensions in distributed deep learning (DL) training. With SPMD, users…

cs.LG2026

S2O: Early Stopping for Sparse Attention via Online Permutation

Yu Zhang, Songwei Liu, Chenqian Yan +4

Attention scales quadratically with sequence length, fundamentally limiting long-context inference. Existing block-granularity sparsification can reduce latency, but coarse blocks…

cs.LG2026

Nemotron 3 Super: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning

NVIDIA, :, Aakshita Chandiramani +544

We describe the pre-training, post-training, and quantization of Nemotron 3 Super, a 120 billion (active 12 billion) parameter hybrid Mamba-Attention Mixture-of-Experts model. Nemo…

cs.DC2026

StaleFlow: Staleness-Aware Data Management for Mitigating Data Skewness in Fully Disaggregated RL Post-Training

Haoyang Li, Sheng Lin, Fangcheng Fu +6

Reinforcement learning (RL) post-training has become pivotal for enhancing the capabilities of modern large models. A recent trend is to develop RL systems with a fully disaggregat…

cs.DC2025

Hydraulis: Balancing Large Transformer Model Training via Co-designing Parallel Strategies and Data Assignment

Haoyang Li, Fangcheng Fu, Sheng Lin +8

To optimize large Transformer model training, both efficient parallel computing and advanced data management are indispensable. However, current methods often assume a stable and u…

cs.DC2025

LobRA: Multi-tenant Fine-tuning over Heterogeneous Data

Sheng Lin, Fangcheng Fu, Haoyang Li +5

With the breakthrough of Transformer-based pre-trained models, the demand for fine-tuning (FT) to adapt the base pre-trained models to downstream applications continues to grow, so…