5 citations · 5 across the 9 of their papers we have counts for
12 papers
PhysBrain 1.5: From Vision-Language Models to Physical Foundation Models
DeepCybo Team, Yu Bin, Haipeng Cao +51
We present PhysBrain 1.5, a unified model for understanding physical environments, generating actions, and predicting future states. Motivated by the physical loop of observation,…
Accelerating Long-Tail Generation in Synchronous RLHF Training via Adaptive Tensor Parallelism
Long Zhao, Qinghe Wang, Jiaan Zhu +5
Reinforcement Learning from Human Feedback (RLHF) has become a key post-training paradigm for improving model quality. However, the synchronous three-stage RLHF pipeline is often b…
Lagom: Unleashing the Power of Communication and Computation Overlapping for Distributed LLM Training
Guanbin Xu, ZhenGuo Xu, Yuzhe Li +4
Overlapping communication with computation is crucial for distributed large-model training, yet optimizing it - especially when computation becomes the bottleneck-remains challengi…
AuroraRL: Fast, Fault-Tolerant, and Cost-Efficient Reinforcement Learning over Decentralized Network
Chaoyi Ruan, Geng Luo, Xinyi Wan +12
LLM reinforcement learning (RL) requires frequent synchronization of large model parameters between the trainer and distributed rollout actors. High-throughput RL post-training the…
Revisiting Parameter Server in LLM Post-Training
Xinyi Wan, Penghui Qi, Guangxing Huang +3
Modern data parallel (DP) training favors collective communication over parameter servers (PS) for its simplicity and efficiency under balanced workloads. However, the balanced wor…
Reaching Agreement Among Reasoning LLM Agents
Chaoyi Ruan, Yiliang Wang, Ziji Shi +1
Multi-agent systems have extended the capability of agentic AI. Instead of single inference passes, multiple agents perform collective reasoning to derive high quality answers. How…
Cortex: Achieving Low-Latency, Cost-Efficient Remote Data Access For LLM via Semantic-Aware Knowledge Caching
Chaoyi Ruan, Chao Bi, Kaiwen Zheng +3
Large Language Model (LLM) agents tackle data-intensive tasks such as deep research and code generation. However, their effectiveness depends on frequent interactions with knowledg…
DynaServe: Unified and Elastic Execution for Dynamic Disaggregated LLM Serving
Chaoyi Ruan, Yinhe Chen, Dongqi Tian +4
LLM inference must meet strict latency SLOs (e.g., 100 ms P99 time-between-tokens) while maximizing goodput. Yet, real-world variability in prompt and response lengths skews comput…
PMPO: Probabilistic Metric Prompt Optimization for Small and Large Language Models
Chenzhuo Zhao, Ziqian Liu, Xinda Wang +2
Prompt optimization is a practical and widely applicable alternative to fine tuning for improving large language model performance. Yet many existing methods evaluate candidate pro…
HCT-QA: A Benchmark for Question Answering on Human-Centric Tables
Mohammad S. Ahmad, Zan A. Naeem, Michaël Aupetit +6
Tabular data embedded in PDF files, web pages, and other types of documents is prevalent in various domains. These tables, which we call human-centric tables (HCTs for short), are…
Fanar: An Arabic-Centric Multimodal Generative AI Platform
Fanar Team, Ummar Abbas, Mohammad Shahmeer Ahmad +39
We present Fanar, a platform for Arabic-centric multimodal generative AI systems, that supports language, speech and image generation tasks. At the heart of Fanar are Fanar Star an…
Hiding Communication Cost in Distributed LLM Training via Micro-batch Co-execution
Haiquan Wang, Chaoyi Ruan, Jia He +4
The growth of Large Language Models (LLMs) has necessitated large-scale distributed training. Highly optimized frameworks, however, still suffer significant losses in Model FLOPS u…