SAPO, Efficient LM Post-Training with Collective RL
This is an academic paper describing SAPO, a meta-algorithm that wraps around your preferred policy gradient algorithm.
Research
CheckFree: fault tolerant training without checkpoints
This is an academic paper describing CheckFree, a novel recovery method for failures in distributed training that does not require checkpointing or redundant computation.
NoLoCo: training large models with no all-reduce
This is an academic paper describing NoLoCo, a novel optimisation method for distributed training that replaces the global synchronisation step with a gossip method.
Diverse Expert Ensembles: embarrassingly parallel LLMs from diverse experts
This is an academic paper that finds benefits to heterogeneity (different model sizes and number of training steps) when training embarrassingly-parallel ensembles of expert models.
SkipPipe: a communication efficient method for decentralised training
This is an academic paper for efficient communication in pipeline parallel training. It introduces an optimal scheduling algorithm that maximises performance and fault tolerance whilst minimising convergence impact from layer skips.
Verde: a verification system for machine learning over untrusted nodes
This is an academic paper describing Verde, a verification protocol for machine learning programs, as well as the underlying Reproducible Operators (RepOps) system that enables it.