arXiv:2606. 14259v1 Announce Type: new Abstract: Prior work has identified several factors that can contribute to the performance gap between Adam and SGD, spanning data aspects, architecture design, and optimization properties.
arXiv:2606. 09658v1 Announce Type: cross Abstract: Muon has recently emerged as a state-of-the-art optimizer for pretraining Large Language Models (LLMs) and vision classifiers.
arXiv:2603. 00742v2 Announce Type: replace Abstract: While Adam has long been the ubiquitous default optimizer for deep neural networks, Muon has recently seen rapid adoption due to its superior training speed.
The paper investigates why the orthogonal optimiser Muon outperforms Adam in large language model pretraining by analysing the spectral properties of Transformer loss landscapes. It finds that Muon’s momentum buffers exhibit an anisotropic spectral profile with a volatile head and a tolerant bulk, enabling larger effective step sizes. Building on this insight, the authors propose Spectral‑Aware Muon (SAMuon) and a lightweight variant, which adjust the bulk scaling while keeping the head unchanged, achieving 13–24 % fewer training tokens than Muon without extra FLOPs.
arXiv:2607. 13246v1 Announce Type: cross Abstract: Muon has recently emerged as a strong optimizer for large-scale deep learning, where it reshapes gradient updates through approximate orthogonalization and has been reported to outperform Adam and AdamW in large language model training.
arXiv:2606. 18587v1 Announce Type: cross Abstract: Decoder-only Transformers compute attention over the KV cache of preceding tokens.
The paper investigates why setting the two momentum parameters of Adam equal (β1=β2) has a special dynamic effect. By analysing Adam in continuous time, the authors show that the update decomposes into a sign component, a magnitude‑lag term proportional to the difference between the two memory times, and other terms. This lag term disappears exactly when β1=β2, making the diagonal the only regime where the mismatch‑induced response is structurally absent. Experiments on six vision and language tasks confirm that tied configurations are sign‑dominated, have smaller lag contributions, and exhibit smoother update‑norm trajectories.
arXiv:2606. 19348v1 Announce Type: cross Abstract: We present a preview version of DeepSeek-V4 series, including two strong Mixture-of-Experts (MoE) language models -- DeepSeek-V4-Pro with 1.
arXiv:2607. 05017v1 Announce Type: cross Abstract: The performance of deep learning models crucially depends on the settings of hyperparameters like learning rate, initialization scale, and weight decay.
We present the first systematic study of Massive activations (MAs) in layer-interleaved HLA LLMs and uncover two architecture-aligned morphologies: MAs consistently spike immediately before full attention layers, forming pre-attention spikes (PAS), and can persist through intervening linear attention layers, giving rise to inter-spike plateaus (ISP). As full attention becomes denser, successive PAS become increasingly connected through ISP, ultimately recovering the stable MA morphology of full attention LLMs.
Muon has recently emerged as a strong optimizer for large-scale deep learning, where it reshapes gradient updates through approximate orthogonalization and has been reported to outperform Adam and AdamW in large language model training. Its empirical success has motivated a growing body of theoretical work that interprets Muon as steepest descent under the spectral norm.
arXiv:2608.12149v2 Announce Type: replace Abstract: We present the first systematic study of Massive activations (MAs) in layer-interleaved HLA LLMs and uncover two architecture-aligned morphologies:...
The paper presents a mean‑field analysis of attention in language models, defining an average attention kernel that propagates representations layer by layer. When conditioned on a whole corpus, the kernel predicts the average evolution of representation geometry; when conditioned on a single context, it predicts the expected geometry for that context. The difference between actual attention and the mean‑field prediction—called the mean‑field deviation—captures context‑specific computation, revealing how models diverge from average behavior during training and in few‑shot tasks.