11 papers
On-Policy Distillation for LLM Safety: A Routing Approach to Template-Robust Realignment
Yongjian Guo, Wanlun Ma, Lingyu Shen +2
Fine-tuning is the dominant paradigm for specializing large language models (LLMs), yet it exposes a critical vulnerability: malicious data providers can embed harmful behaviors in…
JoyNexus: Service-Oriented Multi-Tenant Post-Training for VLA Models
Haoran Sun, Wentao Zhang, Junyang Hua +18
The post-training of Vision-Language-Action (VLA) models is essential due to the diversity of simulators, robot embodiments, and task objectives. Existing compute services, whether…
Embodied Operators and Benchmarking: Toward Reusable and Deployable Embodied Intelligence Systems
Junwu Xiong, Jiaxuan Gao, Wei Chai +10
Embodied intelligence systems require not only end-to-end policy models, but also reusable functional modules that transform multimodal observations, robot states, human demonstrat…
Building a Scalable, Reproducible, Evaluatable, and Closed-Loop Simulation Environment Foundation for Embodied Intelligence
Junwu Xiong, Yongjian Guo, Mingxi Luo +17
This paper presents a cloud-native simulation infrastructure framework for embodied intelligence that supports large-scale training, standardized evaluation, and simulation-based d…
AdaptiveLoad: Towards Efficient Video Diffusion Transformer Training
Yucheng Guo, Yongjian Guo, Zhong Guan +6
In video generation models, particularly world models, training large-scale video diffusion Transformers (such as DiT and MMDiT) poses significant computational challenges due to t…
Missing Old Logits in Asynchronous Agentic RL: Semantic Mismatch and Repair Methods for Off-Policy Correction
Zhong Guan, Yongjian Guo, Haoran Sun +5
Asynchronous reinforcement learning improves rollout throughput for large language model agents by decoupling sample generation from policy optimization, but it also introduces a c…