Allocating Recurrent Compute in Looped Language Models
cs.LG
Submitted: 2026-08-18
Updated: 2026-11-12
Terminology
Sources
- Universal Transformers
- Looped Transformers are Better at Learning Learning Algorithms
- Reasoning with Latent Thoughts: On the Power of Looped Transformers
- Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach
- Scaling Latent Reasoning via Looped Language Models
- LT2: Linear-Time Looped Transformers
- Memory-Efficient Looped Transformer: Decoupling Compute from Memory in Looped Language Models
- Depth-Recurrent Attention Mixture (Dreamer): Giving Latent Reasoning the Attention it Deserves
- Skip a Layer or Loop it? Test-Time Depth Adaptation of Pretrained LLMs
- LoopFormer: Elastic-Depth Looped Transformers for Latent Reasoning via Shortcut Modulation
- Think-at-Hard: Dynamic Looped Transformers for Improved Reasoning
- Sparse Layers are Critical to Scaling Looped Language Models
- LoopMoE: Unifying Iterative Computation with Mixture-of-Experts for Language Modeling
- Linear Transformers Are Secretly Fast Weight Programmers
- Gated Delta Networks: Improving Mamba2 with Delta Rule
- Gated DeltaNet-2: Decoupling Erase and Write in Linear Attention
- DataComp-LM: In search of the next generation of training sets for language models
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