Results for “stm”

Submit and manage TAO training, evaluation, and inference jobs on SLURM GPU clusters over SSH with sbatch/srun, Pyxis/Enroot containers, and Lustre-backed storage.

Launch distributed Megatron-LM training jobs on a SLURM cluster with a minimal sbatch skeleton, environment-variable setup for torch.distributed.run, CUDA_DEVICE_MAX_CONNECTIONS rules, container conventions, monitoring, and per-rank failure diagnosis.

Quantifies audio source separation quality by computing the signal-to-distortion ratio (SDR) between ground-truth and estimated stems, with per-stem and record-level averaging.

Post-train LLMs with reinforcement learning using the slime framework, which integrates Megatron-LM for training and SGLang for rollout generation.

Train and run Mamba state-space models with O(n) complexity, achieving faster inference and longer context than Transformers.

Evaluates spatial realism and temporal flow consistency of AI-generated videos using embedding spaces and Fourier transforms, producing bounded STREAM-S and STREAM-T scores.