Shared Weights, Selected Computations: How Looped Transformers Route What Each Loop Does
cs.LG
Submitted: 2026-09-30
Updated: 2026-09-30
Code: https://github.com/wjjpku/howloop
Terminology
Sources
- A Mechanistic Analysis of Looped Reasoning Language Models
- Universal Transformers
- Bridging the Gap Between Latent and Explicit Reasoning with Looped Transformers
- Adaptive Depth in Looped Transformers: Diagnosing Learned Halting Gates and Trajectory Readouts
- Reasoning with Latent Thoughts: On the Power of Looped Transformers
- Dense Supervision Is Not Enough: The Readout Blind Spot in Looped Language Models
- Steering Language Models With Activation Engineering
- Attention Is All You Need
- When Does Recurrence Become an Algorithm? Convergence Selection in Weight-Tied Looped Transformers
- Scaling Latent Reasoning via Looped Language Models
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks