arXiv AI September 28, 2026 By Justus Westerhoff, Stephan Olbrich, Hatem Oraby, Matthew Evan Larkum, Felix Alexander Gers

Programs-of-Layers in LLMs through the Lens of Cortical Areas

Read the original on arXiv AI →

The paper examines a method called Program-of-Layers (PoLar) that allows transformer layers to be dynamically routed rather than processed in a fixed sequence, mirroring the brain’s thalamic routing. Reproductions across five models confirm that skipping, repeating, and combining layer blocks improve performance, with shorter programs for easier inputs and more repeats for harder ones. However, the study could not replicate the claimed advantage of a learned single‑shot router, noting that its top prediction defaults to the standard pass while the top‑k predictions still yield accuracy gains. The authors also analyze the robustness of correction programs, finding them brittle to single edits, and release their code publicly.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

llms

arXiv Machine Learning

Jun 8

Skip a Layer or Loop It? Learning Program-of-Layers in LLMs

arXiv:2606. 06574v1 Announce Type: new Abstract: Large language models (LLMs) perform inference by following a fixed depth and order, non-recurrent execution of all layers.

By Ziyue Li, Yang Li, Tianyi Zhou

llms benchmarks

arXiv AI

Sep 16

Thinking Deeper, Not Longer: Memory-Efficient Test-Time Reasoning with Depth-Recurrent Transformers for Compositional Generalization

arXiv:2603.21676v2 Announce Type: replace-cross Abstract: Standard Transformers have a fixed computational depth, limiting their ability to generalize to tasks that require variable-depth reasoning....

By Hung-Hsuan Chen

llms safety

arXiv AI

Jul 8

Transformers converge to invariant algorithmic cores

arXiv:2602. 22600v2 Announce Type: replace-cross Abstract: Training selects for behavior, not circuitry: many weight configurations can implement the same function.

By Joshua S. Schiffman

llms

arXiv Machine Learning

Aug 31

Think-at-Hard: Dynamic Looped Transformers for Improved Reasoning

arXiv:2511.08577v4 Announce Type: replace-cross Abstract: Improving the reasoning abilities of Large Language Models (LLMs), especially under parameter constraints, is crucial for real-world applicat...

By Tianyu Fu, Yichen You, Zekai Chen, Guohao Dai, Huazhong Yang, Yu Wang

llms fine-tuning benchmarks

arXiv AI

Sep 2

Flow Reasoning Models: Turning Flows Into Efficient Recurrent Reasoners

Flow Reasoning Models (FRMs) are a new framework that turns continuous flow models into efficient recurrent reasoners for structured tasks. By self‑conditioning a flow model on its own past outputs, FRMs iteratively refine solutions, allowing parallel decision making and revision. The authors introduce Fixed‑Point Forcing (FPF) to mitigate exposure bias at deeper recursion, and report near‑perfect solve rates on Sudoku‑Extreme, Zebra, and Maze‑Unique, outperforming existing masked‑diffusion and specialized baselines while using far fewer inference FLOPs.

By Alec Helbling, Andrey Bryutkin, Mauro Martino, Duen Horng Chau, Nima Dehmamy, Hendrik Strobelt

diffusion safety

arXiv AI

Jun 16

Think-at-Hard: Selective Latent Iterations to Improve Reasoning Language Models

arXiv:2511. 08577v3 Announce Type: replace-cross Abstract: Improving the reasoning abilities of Large Language Models (LLMs), especially under parameter constraints, is crucial for real-world applications.

By Tianyu Fu, Yichen You, Zekai Chen, Guohao Dai, Huazhong Yang, Yu Wang

llms fine-tuning benchmarks

One email a morning, machine-written

Related stories

Skip a Layer or Loop It? Learning Program-of-Layers in LLMs

Thinking Deeper, Not Longer: Memory-Efficient Test-Time Reasoning with Depth-Recurrent Transformers for Compositional Generalization

Transformers converge to invariant algorithmic cores

Think-at-Hard: Dynamic Looped Transformers for Improved Reasoning

Flow Reasoning Models: Turning Flows Into Efficient Recurrent Reasoners

Think-at-Hard: Selective Latent Iterations to Improve Reasoning Language Models