Looping Beyond Twice: A Scalable Recipe for Looped Mixture-of-Experts
cs.LG, cs.AI
Submitted: 2026-10-01
Updated: 2026-10-01
Code: https://github.com/hed-ucas/LOOM
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- PIQA: Reasoning about Physical Commonsense in Natural Language
- LoopMoE: Unifying Iterative Computation with Mixture-of-Experts for Language Modeling
- Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
- Universal Transformers
- Simply Stabilizing the Loop via Fully Looped Transformer
- Loop the Loopies!
- Training Large Language Models to Reason in a Continuous Latent Space
- Training Compute-Optimal Large Language Models
- MiniCPM: Unveiling the Potential of Small Language Models with Scalable Training Strategies
- Mixtral of Experts
- Less is More: Recursive Reasoning with Tiny Networks
- Scaling Laws for Neural Language Models
- Nanbeige4.2-3B: Unlocking Agentic Capabilities in a Compact Model
- ALBERT: A Lite BERT for Self-supervised Learning of Language Representations
- Sparse Layers are Critical to Scaling Looped Language Models
- Mix-LN: Unleashing the Power of Deeper Layers by Combining Pre-LN and Post-LN
- Decoupled Weight Decay Regularization
- Fixed-Point Reasoners: Stable and Adaptive Deep Looped Transformers
- GLU Variants Improve Transformer
- Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer
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