# What Makes Looped Transformers Perform Better Than Non-Recursive Ones

Authors: Zixuan Gong, Yong Liu, Jiaye Teng.

Studies how repeated attention changes optimization through a model of loss-landscape geometry. Proposes SHIFT, which starts with one attention pass and adds shared-weight repetitions after learning stalls, and tests the schedule on synthetic sequence tasks.

First public: 2025-10-11.
ArXiv initial submission: 2025-10-11.

[Original paper](https://arxiv.org/abs/2510.10089) · [PDF](https://arxiv.org/pdf/2510.10089) · [HTML](https://arxiv.org/html/2510.10089v3) · [Reading-list entry](https://latent-implicit-thinking.github.io/papers/2510.10089/)
[Preprint BibTeX](https://latent-implicit-thinking.github.io/papers/2510.10089/citation.bib) · [Preprint CSL-JSON](https://latent-implicit-thinking.github.io/papers/2510.10089/citation.csl.json) · [JSON record](https://latent-implicit-thinking.github.io/papers/2510.10089/record.json)

## Approaches

- [Recurrent reasoning](https://latent-implicit-thinking.github.io/topics/recurrent-reasoning/): Repeat parts of the model or keep intermediate information in memory while working through a problem. Some methods also learn how many passes to run.
- [Analysis and evaluation](https://latent-implicit-thinking.github.io/topics/implicit-reasoning-evaluation/): Measure how well these methods work, investigate what their internal computation does, and identify where they fail. This includes experiments, studies of model internals, and mathematical results.

Reading-list version 0.15.2; literature checked through 2026-11-03.