Propagation of Chaos in Contextual Flow Maps
cs.LG, math.AP, math.OC, math.PR, math.ST, stat.TH
Submitted: 2026-05-16
Updated: 2026-10-30
Comments: 32 pages, 1 figure
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Attention's forward pass and Frank-Wolfe
- Perceptrons and localization of attention's mean-field landscape
- Softmax as Linear Attention in the Large-Prompt Regime: a Measure-based Perspective
- Scaling Limits of Long-Context Transformers
- Token Sample Complexity of Attention
- Multi-layer Cross-attention is Provably Optimal for Multi-modal In-context Learning
- A Unified Perspective on the Dynamics of Deep Transformers
- ResNets of All Shapes and Sizes: Convergence of Training Dynamics in the Large-scale Limit
- The Hidden Width of Deep ResNets: Tight Error Bounds and Phase Diagram
- Quantitative Clustering in Mean-Field Transformer Models
- Approximation Theory for Lipschitz Continuous Transformers
- Dynamic metastability in the self-attention model
- Uniform-in-time concentration in two-layer neural networks via transportation inequalities
- Measure-to-measure interpolation using Transformers
- Propagation of Chaos in One-hidden-layer Neural Networks beyond Logarithmic Time
- Transformers as Measure-Theoretic Associative Memory: A Statistical Perspective and Minimax Optimality
- The Mean-Field Dynamics of Transformers
- Separating Geometry from Probability in the Analysis of Generalization
- Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context
- A Mathematical Theory of Attention
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis