English
Related papers

Related papers: Attention's forward pass and Frank-Wolfe

200 papers

We study dynamical scaling in the quantum-critical fan of the pseudogap-metal to Fermi-liquid transition of the two-dimensional Hubbard model. Using a four-patch dynamical cluster approximation with the numerical renormalization group as a…

Strongly Correlated Electrons · Physics 2026-05-15 Mathias Pelz , Gabriel Kotliar , Jan von Delft , Andreas Gleis

Deploying Transformer models on edge devices is limited by latency and energy budgets. While INT8 quantization effectively accelerates the primary matrix multiplications, it exposes the softmax as the dominant bottleneck. This stage incurs…

Machine Learning · Computer Science 2025-11-27 Wanli Zhong , Haibo Feng , Zirui Zhou , Hanyang Peng , Shiqi Yu

Large models based on the Transformer architecture are susceptible to extreme-token phenomena, such as attention sinks and value-state drains. These issues, which degrade model performance, quantization fidelity, and interpretability, arise…

Machine Learning · Computer Science 2026-01-27 Rui Bu , Haofeng Zhong , Wenzheng Chen , Yangyan Li

The softmax function is crucial in Transformer attention, which normalizes each row of the attention scores with summation to one, achieving superior performances over other alternative functions. However, the softmax function can face a…

Computation and Language · Computer Science 2025-02-26 Chuanyang Zheng , Yihang Gao , Guoxuan Chen , Han Shi , Jing Xiong , Xiaozhe Ren , Chao Huang , Xin Jiang , Zhenguo Li , Yu Li

Vision transformers using self-attention or its proposed alternatives have demonstrated promising results in many image related tasks. However, the underpinning inductive bias of attention is not well understood. To address this issue, this…

Machine Learning · Computer Science 2022-05-23 Arda Sahiner , Tolga Ergen , Batu Ozturkler , John Pauly , Morteza Mardani , Mert Pilanci

The Frank-Wolfe method (a.k.a. conditional gradient algorithm) for smooth optimization has regained much interest in recent years in the context of large scale optimization and machine learning. A key advantage of the method is that it…

Optimization and Control · Mathematics 2015-08-17 Dan Garber , Elad Hazan

We examine the intrinsic (within the attention head) and extrinsic (amongst the attention heads) structure of the self-attention mechanism in transformers. Theoretical evidence for invariance of the self-attention mechanism to softmax…

Numerical Analysis · Mathematics 2025-06-19 Oluwadamilola Fasina , Ruben V. C. Pohle , Pei-Chun Su , Ronald R. Coifman

Zero-range processes with decreasing jump rates are known to exhibit condensation, where a finite fraction of all particles concentrates on a single lattice site when the total density exceeds a critical value. We study such a process on a…

Probability · Mathematics 2018-04-26 Inés Armendáriz , Stefan Grosskinsky , Michail Loulakis

Standard transformer architectures apply a single attention mechanism uniformly across all tokens and sequence positions, irrespective of local context or computational budget. We propose Meta-Attention, a framework that dynamically routes…

Machine Learning · Computer Science 2026-05-28 Alan Ferrari

Transformer-based models have emerged as one of the most widely used architectures for natural language processing, natural language generation, and image generation. The size of the state-of-the-art models has increased steadily reaching…

Hardware Architecture · Computer Science 2025-01-15 Rya Sanovar , Srikant Bharadwaj , Renee St. Amant , Victor Rühle , Saravan Rajmohan

Memory is a key computational bottleneck when solving large-scale convex optimization problems such as semidefinite programs (SDPs). In this paper, we focus on the regime in which storing an $n\times n$ matrix decision variable is…

Optimization and Control · Mathematics 2021-08-25 Nimita Shinde , Vishnu Narayanan , James Saunderson

Softmax-based dot-product attention is a cornerstone of Transformer architectures, enabling remarkable capabilities such as in-context learning. However, as context lengths increase, a fundamental limitation of the softmax function emerges:…

Machine Learning · Computer Science 2026-02-12 Sai Surya Duvvuri , Nirmal Patel , Nilesh Gupta , Inderjit S. Dhillon

Despite powering modern AI, transformers remain mysteriously brittle to train. We develop a stability theory that explains why pre-LayerNorm works, why DeepNorm uses $N^{-1/4}$ scaling, and why warmup is necessary, all from first…

Machine Learning · Computer Science 2026-02-24 Seyed Morteza Emadi

We study how multi-head softmax attention models are trained to perform in-context learning on linear data. Through extensive empirical experiments and rigorous theoretical analysis, we demystify the emergence of elegant attention patterns:…

Machine Learning · Computer Science 2025-05-29 Jianliang He , Xintian Pan , Siyu Chen , Zhuoran Yang

Transformers have recently revolutionized many domains in modern machine learning and one salient discovery is their remarkable in-context learning capability, where models can solve an unseen task by utilizing task-specific prompts without…

Machine Learning · Computer Science 2023-10-10 Yu Huang , Yuan Cheng , Yingbin Liang

Dynamical quantum-cluster approaches, such as different cluster extensions of the dynamical mean-field theory (cluster DMFT) or the variational cluster approximation (VCA), combined with efficient cluster solvers, such as the quantum…

Strongly Correlated Electrons · Physics 2013-05-29 Gang Li , Werner Hanke , Alexei N. Rubtsov , Sebastian Bäse , Michael Potthoff

Attention is a key part of the transformer architecture. It is a sequence-to-sequence mapping that transforms each sequence element into a weighted sum of values. The weights are typically obtained as the softmax of dot products between…

FlashAttention-style online softmax enables exact attention computation with linear memory by streaming score tiles through on-chip memory and maintaining a running maximum and normalizer. However, as attention kernels approach peak…

Machine Learning · Computer Science 2026-04-15 Yupeng Sun , Yanzhao Li , Zhiqiang Zou , Bai Du , Zhiyuan Zhang , Hui Dong , Gaoyige Fan , Hui Wang

Frozen Density Embedding Theory (FDET) [Wesolowski {\it Phys. Rev. A} {\bf 77}, 012504 (2008)] provides the interpretation of the eigenvalue equations for an embedded $N'$-electron wavefunction, in which the embedding operator is…

Chemical Physics · Physics 2025-07-02 Tomasz Adam Wesolowski

Large language models spend most of their inference cost on attention over long contexts, yet empirical behavior suggests that only a small subset of tokens meaningfully contributes to each query. We formalize this phenomenon by modeling…

Artificial Intelligence · Computer Science 2026-02-17 Vashista Nobaub