5 papers
MARLaaS: Multi-Tenant Asynchronous Reinforcement Learning as a Service
Timothy Tin Long Yu, Gursimran Singh, Ge Shi +3
Reinforcement Learning from Verifiable Rewards (RLVR) has significantly improved the reasoning capabilities of large language models (LLMs), particularly in multi-turn agentic sett…
ElasticMoE: An Efficient Auto Scaling Method for Mixture-of-Experts Models
Gursimran Singh, Timothy Yu, Haley Li +7
Mixture-of-Experts (MoE) models promise efficient scaling of large language models (LLMs) by activating only a small subset of experts per token, but their parallelized inference p…
Efficiently Serving Large Multimodal Models Using EPD Disaggregation
Gursimran Singh, Xinglu Wang, Yifan Hu +9
Large Multimodal Models (LMMs) extend Large Language Models (LLMs) by handling diverse inputs such as images, audio, and video, but at the cost of adding a multimodal encoding stag…
Enhancing Learned Knowledge in LoRA Adapters Through Efficient Contrastive Decoding on Ascend NPUs
Morgan Lindsay Heisler, Linzi Xing, Ge Shi +7
Huawei Cloud users leverage LoRA (Low-Rank Adaptation) as an efficient and scalable method to fine-tune and customize large language models (LLMs) for application-specific needs. H…
DivPrune: Diversity-based Visual Token Pruning for Large Multimodal Models
Saeed Ranjbar Alvar, Gursimran Singh, Mohammad Akbari +1
Large Multimodal Models (LMMs) have emerged as powerful models capable of understanding various data modalities, including text, images, and videos. LMMs encode both text and visua…