5 papers
F-TIS: Harnessing Diverse Models in Collaborative GRPO
Nikolay Blagoev, OÄuzhan Ersoy, Wendelin Boehmer +1
Reinforcement learning methods such as GRPO have seen great popularity in LLM post-training. In GRPO, models produce completions to a set of prompts, which are rewarded, and the po…
Training-Free Dynamic Upcycling of Expert Language Models
Eros Fanì, OÄuzhan Ersoy
Large Language Models (LLMs) have achieved remarkable performance on a wide range of specialized tasks, exhibiting strong problem-solving capabilities. However, training these mode…
NoLoCo: No-all-reduce Low Communication Training Method for Large Models
Jari Kolehmainen, Nikolay Blagoev, John Donaghy +2
Training large language models is generally done via optimization methods on clusters containing tens of thousands of accelerators, communicating over a high-bandwidth interconnect…
SkipPipe: Partial and Reordered Pipelining Framework for Training LLMs in Heterogeneous Networks
Nikolay Blagoev, Lydia Yiyu Chen, OÄuzhan Ersoy
Data and pipeline parallelism are ubiquitous for training of Large Language Models (LLM) on distributed nodes. Driven by the need for cost-effective training, recent work explores…
HDEE: Heterogeneous Domain Expert Ensemble
OÄuzhan Ersoy, Jari Kolehmainen, Gabriel Passamani Andrade
Training dense LLMs requires enormous amounts of data and centralized compute, which introduces fundamental bottlenecks and ever-growing costs for large models. Several studies aim…