LoopICL: Looping a single transformer block to solve tabular tasks

arXiv:2609.36108 · cs.LG, cs.AI · Submitted 2026-09-28 · Read on arXiv

cs.LG, cs.AI

Submitted: 2026-09-28

Updated: 2026-09-28

Code: https://github.com/amirbalef/loopicl

Terminology

• Transformer — 1× in this paper · explained in 2 episodes

Sources

• A Mechanistic Analysis of Looped Reasoning Language Models
• EXAONE Tabular 1.0 : Technical Report
• ELT: Elastic Looped Transformers for Visual Generation
• Adaptive Computation Time for Recurrent Neural Networks
• TabPFN-3: Technical Report
• Less is More: Recursive Reasoning with Tiny Networks
• Test-Time Compute Scaling for ASR with Depth-Conditioned Looped Transformers
• State-Space Models for Tabular Prior-Data Fitted Networks
• TACTICL: Task-Aware Compression of Tabular ICL Models
• ScoringBench: A Benchmark for Evaluating Tabular Foundation Models with Proper Scoring Rules
• Fixed-Point Reasoners: Stable and Adaptive Deep Looped Transformers
• Looped Latent Attention: Cross-Loop KV Compression for Looped Transformers
• Tab-TRM: Tiny Recursive Model for Insurance Pricing on Tabular Data
• TabH2O: A Unified Foundation Model for Tabular Prediction
• TabICLv2: A better, faster, scalable, and open tabular foundation model
• Mambular: A Sequential Model for Tabular Deep Learning
• Memory-Efficient Looped Transformer: Decoupling Compute from Memory in Looped Language Models
• On the Residual Scaling of Looped Transformers: Stability and Transferability
• Stabilizing Recurrent Dynamics for Test-Time Scalable Latent Reasoning in 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