[2606.04366] MeshTok: Efficient Multi-Scale Tokenization for Scalable PDE Transformers
Abstract page for arXiv paper 2606.04366: MeshTok: Efficient Multi-Scale Tokenization for Scalable PDE Transformers
America Forever Bytes
Technology
Abstract page for arXiv paper 2606.04366: MeshTok: Efficient Multi-Scale Tokenization for Scalable PDE Transformers
Abstract page for arXiv paper 2606.04032: Do Transformers Need Three Projections? Systematic Study of QKV Variants
Abstract page for arXiv paper 2606.04405: Low-Rank Decay for Grokking in Scale-Invariant Transformers: A Spectral-Geometric View
Abstract page for arXiv paper 2606.04752: An Empirical Audit of Input Encoders for Multi-Channel Signal Transformers
Abstract page for arXiv paper 2505.17315: Longer Context, Deeper Thinking: Uncovering the Role of Long-Context Ability in Reasoning
Abstract page for arXiv paper 2510.24342: A Unified Geometric Space for Topological Alignment Between Transformer-Based Models and Human Brain Networks
Abstract page for arXiv paper 2606.00114: Recursive Vision Transformer with Dynamic Depth and Width Adjustment for Resource-Efficient Image Semantic Communicati...
Abstract page for arXiv paper 2511.16886: Latent Reasoning in TRMs is Secretly a Policy Improvement Operator
Abstract page for arXiv paper 2606.01532: Rethinking the Role of Positional Encoding: Sliding-Window Transformers without PE Remain Turing Complete
Abstract page for arXiv paper 2605.31204: Probabilistic Precipitation Nowcasting with Rectified Flow Transformers
Abstract page for arXiv paper 2605.31367: Trading Complexity for Expressivity Through Structured Generalized Linear Token Mixing
Abstract page for arXiv paper 2605.30523: Revisiting Padded Transformer Expressivity: Which Architectural Choices Matter and Which Don't
Abstract page for arXiv paper 2605.29634: Relational Rank Geometry in Transformers: Detecting and Steering Hidden-State Relation Frames
Abstract page for arXiv paper 2605.29754: Benchmarking Positional Encoding Strategies for Transformer-Based EEG Foundation Models
The Transformers franchise continues to set new records with its merchandise and collectibles, and this latest sale is hard to beat.
Abstract page for arXiv paper 2605.27458: Generic Interpretation Approach for Transformer Models Incorporating Heterogenous Attention Structures
Abstract page for arXiv paper 2605.28592: PLS in the Mirror of Self-Attention
Abstract page for arXiv paper 2605.28600: Transformers Provably Learn to Internalize Chain-of-Thought
Abstract page for arXiv paper 2605.28075: Measure-to-measure Regression with Transformers