arXiv:2607. 01678v1 Announce Type: new Abstract: Communication increasingly dominates the cost of Large Language Model (LLM) pre-training, especially under data-parallel and sharded training schemes, where gradient synchronization and parameter reconstruction overhead increase with model size and system scale.
By Mingkai Zheng, Junlin Chen, Haotian Xie, Zhao Zhang
The paper introduces Primus Projection, a hybrid engine that combines GPU microbenchmarks with analytical scaling to predict multi‑node distributed training performance, and Primus Tuning Agent, an automated search that uses the projection engine as a fast oracle to explore the vast configuration space of parallelism and micro‑batching. Together, they reduce the need for costly trial‑and‑error on hundreds of GPUs, achieving predictions within ~10% of real throughput and enabling significant throughput gains—up to +27% in a Mixtral 8x22B case study—within minutes of single‑node exploration.
By Anshu Raina, Peyman Razaghi, Yuankai Chen, Cheng Yao, Yao Fu, Vidushi Goyal, Devang Patel, Wen Xie, Zhenyu Gu, Emad Barsoum
arXiv:2504. 15610v4 Announce Type: replace Abstract: Fine-tuning a 7B language model for specialized advising is attractive in resource-constrained settings, but multi-epoch runs routinely exceed the wall-clock limits of the free-tier GPUs (Kaggle, Colab) such users rely on.
By Md Millat Hosen
Configuring distributed LLM training across hundreds of GPUs requires jointly optimizing tensor, pipeline, expert, context, and data parallelism alongside micro-batch size, recomputation, sharding, an...
arXiv:2609.38095v1 Announce Type: new
Abstract: Backpropagation (BP) dominates deep learning but imposes a massive memory tax. For example, training OPT-30B with Adam requires $\approx$ 600GB of GPU...
By Francois Chaubard, Mykel J. Kochenderfer, Chris R\'e
arXiv:2502. 11034v3 Announce Type: replace Abstract: Loss spikes remain a persistent obstacle in large-scale language model pretraining.
By Guoxia Wang, Shuai Li, Congliang Chen, Jinle Zeng, Jiabin Yang, Dianhai Yu, Yanjun Ma, Li Shen
arXiv:2608. 03796v1 Announce Type: cross Abstract: Small language models are often the only option for deployment under tight latency, cost, and on-premises constraints, but they are rarely trained from scratch: a compressed model is usually recovered through knowledge distillation (KD).
By Bakbergen Ryskulov, Iker Garc\'ia-Ferrero, David Montero, David Jansen, Ali Hashemi, Jezabel R. Garcia, Antonio Tiene, Rom\'an Or\'us
arXiv:2607. 16612v1 Announce Type: cross Abstract: Backpropagation makes training deep networks memory intensive because it must store intermediate activations.
By Tian Qin, Wei-Min Huang
arXiv:2606. 24722v1 Announce Type: new Abstract: Frontier AI training is increasingly shaped by access to dense, centrally controlled accelerator clusters.
By Peter Toth
Serving an LLM efficiently depends on the engine, how the model is split across GPUs, precision, scheduler settings, speculative decoding, KV-cache management, and prefill-decode disaggregation. The b...
Model Annealing is a data‑efficient learning method that, after an initial memorization phase, spends the remaining training budget on review and consolidation. The training is split into episodes with cyclic learning rates; each episode learns from true labels, tempered predictions stored in a memory of past models, and adversarial ‘dreams’ where the student and memory disagree most. Across several datasets and architectures, Model Annealing improves top‑1 accuracy, likelihood, calibration, and robustness compared to standard cross‑entropy training and matches snapshot distillation while adding no inference overhead.
By Ali Mohammadi Ruzbahani, Rezvan Abouzadeh
The paper investigates training‑free pruning of Mixture‑of‑Experts (MoE) models, focusing on how to handle the computation when a router selects a pruned expert. It compares replacement operators from three families on several models (OLMoE, DeepSeek‑V2‑Lite, Moonlight, Granite) while keeping the pruning mask fixed, and introduces a new operator called the tilted novice. The study finds that no single operator consistently dominates, with performance varying by model, pruning level, and evaluation metric, and it proposes a five‑item checklist for reporting results.
By Ezza Ali, Sehrish Bibi, Momina Moetesum, Faisal Shafait