5 papers
Hera: Learning Long-Horizon Coordination for Device-Cloud Collaborative LLM Agents
Yuxin Zhang, Mengxue Hu, Zheng Lin +8
Large language model (LLM) agents excel at solving complex long-horizon tasks through autonomous interaction with environments. However, their real-world deployment faces a fundame…
DyGRO-VLA: Cross-Task Scaling of Vision-Language-Action Models via Dynamic Grouped Residual Optimization
Sixu Lin, Yunpeng Qing, Litao Liu +4
Recent progress in Reinforcement Learning (RL) provides a principled approach to optimizing Vision-Language-Action (VLA) models, facilitating a shift from trajectory imitation to a…
GoodServe: Towards High-Goodput Serving of Agentic LLM Inferences over Heterogeneous Resources
Boxiao Du, Boning Huangfu, Yizhou Luo +5
Large Language Models (LLMs) play a critical role in emerging agentic applications, where the timely completion of each entire inference is critical. Meanwhile, agentic LLM inferen…
S-HPLB: Efficient LLM Attention Serving via Sparsity-Aware Head Parallelism Load Balance
Di Liu, Yifei Liu, Chen Chen +4
With the increasing volumes of Large Language Models (LLMs) and the expanding context lengths, attention computation has become a key performance bottleneck in LLM serving. For fas…
LLMSched: Uncertainty-Aware Workload Scheduling for Compound LLM Applications
Botao Zhu, Chen Chen, Xiaoyi Fan +1
Developing compound Large Language Model (LLM) applications is becoming an increasingly prevalent approach to solving real-world problems. In these applications, an LLM collaborate…