LLMs: architectures, context, training — page 2

• Latent Cache Flow: Model-to-Model Communication Without Text
  arxiv:2605.22863 · 2026-05-19 · text available
• Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention
  arxiv:2404.07143 · 2024-04-10 · text available
• LLaDA2.0: Scaling Up Diffusion Language Models to 100B
  arxiv:2512.15745 · 2025-12-10 · text available
• LLaDA2.1: Speeding Up Text Diffusion via Token Editing
  arxiv:2602.08676 · 2026-02-09 · text available
• Memorizing Transformers
  arxiv:2203.08913 · 2022-03-16 · text available
• Mixture-of-Recursions: Learning Dynamic Recursive Depths for Adaptive Token-Level Computation
  arxiv:2507.10524 · 2025-07-14 · text available
• Neural Networks and the Chomsky Hierarchy
  arxiv:2207.02098 · 2022-07-05 · text available
• Olmo 3
  arxiv:2512.13961 · 2025-12-15 · text available
• Recirculation
  arxiv:2608.17981 · 2026-08-18 · text available
• Recursive Language Models
  arxiv:2512.24601 · 2025-12-31 · text available
• Scaling MLPs: A Tale of Inductive Bias
  arxiv:2306.13575 · 2023-06-23 · text available
• Searching for Activation Functions
  arxiv:1710.05941 · 2017-10-16 · text available
• Skip a Layer or Loop It? Learning Program-of-Layers in LLMs
  arxiv:2606.06574 · 2026-06-04 · text available
• Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference
  arxiv:2412.13663 · 2024-12-18 · text available
• T5Gemma 2: Seeing, Reading, and Understanding Longer
  arxiv:2512.14856 · 2025-12-16 · text available
• The Llama 3 Herd of Models
  arxiv:2407.21783 · 2024-07-31 · text available
• The most cited neural nets of the 20th century (Schmidhuber)
  url:people.idsia.ch/~juergen/most-cited-neural-nets.html · 2021-12-01 · metadata only
• The Topological Trouble With Transformers
  arxiv:2604.17121 · 2026-04-18 · text available
• TransformerFAM: Feedback attention is working memory
  arxiv:2404.09173 · 2024-04-14 · text available
• Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention
  arxiv:2006.16236 · 2020-06-29 · text available
• Universal Transformers
  arxiv:1807.03819 · 2018-07-10 · text available
• Unlimiformer: Long-Range Transformers with Unlimited Length Input
  arxiv:2305.01625 · 2023-05-02 · text available
• XBridge: Entity-Grounded Latent Bridge for Heterogeneous LLM Communication
  arxiv:2608.11676 · 2026-08-12 · text available
• xLSTM: Extended Long Short-Term Memory
  arxiv:2405.04517 · 2024-05-07 · text available