6 papers
d-OPD: Future-Aware On-Policy Distillation for Block Diffusion Language Models
Ruitao Liu, Qinghao Hu, Song Han
Large language models (LLMs) typically generate text autoregressively (AR), predicting one token at a time. Block diffusion language models (dLLMs) instead generate blocks sequenti…
Hide to Guide: Learning via Semantic Masking
Ruitao Liu, Qinghao Hu, Alex Hu +6
Reinforcement learning with verifiable rewards (RLVR) has become a powerful paradigm for improving language models on reasoning-intensive tasks, but its effectiveness is often limi…
A Readiness-Driven Runtime for Pipeline-Parallel Training under Runtime Variability
Ruitao Liu, Xinyang Tian, Shuo Chen +4
Pipeline parallelism is a key technique for scaling large-model training, but modern workloads exhibit runtime variability in computation and communication. Existing pipeline syste…
FFTrainer: Fast Failover in Large-Language Model Training with Almost-Free State Management
Bohan Zhao, Yuanhong Wang, Chenglin Liu +6
Recent developments in large language models (LLMs) have introduced new requirements for efficient and robust training. As LLM clusters scale, node failures, lengthy recoveries, an…
MegatronApp: Efficient and Comprehensive Management on Distributed LLM Training
Bohan Zhao, Guang Yang, Shuo Chen +4
The rapid escalation in the parameter count of large language models (LLMs) has transformed model training from a single-node endeavor into a highly intricate, cross-node activity.…
SyncDiff: Synchronized Motion Diffusion for Multi-Body Human-Object Interaction Synthesis
Wenkun He, Yun Liu, Ruitao Liu +1
Synthesizing realistic human-object interaction motions is a critical problem in VR/AR and human animation. Unlike the commonly studied scenarios involving a single human or hand i…