#fine-tuning
Every summary, chronological. Filter by category, tag, or source from the rail.
ATOD: Hybrid Distillation for Autonomous Agent Training
ATOD combines on-policy distillation with reinforcement learning using an annealed schedule and turn-level reweighting to train small agent models that outperform their larger teacher models.
Fine-Tuning Tiny LLMs for On-Device AI Agents
Developers can achieve production-grade performance on-device by choosing between system-level models (Gemini Nano) for general tasks or fine-tuning tiny LLMs (<1B parameters) via LiteRT-LM for specialized, high-accuracy agentic workflows.
AI EngineerAccelerating MoE Fine-Tuning with NVIDIA NeMo AutoModel
NVIDIA NeMo AutoModel extends Hugging Face Transformers v5 to provide 3.4-3.7x higher training throughput and 29-32% lower memory usage for MoE models by integrating Expert Parallelism, DeepEP, and TransformerEngine kernels.
Showing 3 of 3