# Looped Transformers under the Jacobian Lens: Does the Global Workspace Survive Recurrence?

arXiv: https://arxiv.org/abs/2609.01924 · HTML: https://mlsanity.com/paper/2609.01924/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 20 references that are in the map.
- First posted on arXiv 1 Sep 2026.

## Builds on (most cited first, 11 of 11)

- [ALBERT: A Lite BERT for Self-supervised Learning of Language Representations](https://mlsanity.com/paper/1909.11942/) (2019)
- [Universal Transformers](https://mlsanity.com/paper/1807.03819/) (2018)
- [Eliciting Latent Predictions from Transformers with the Tuned Lens](https://mlsanity.com/paper/2303.08112/) (2023)
- [Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach](https://mlsanity.com/paper/2502.05171/) (2025)
- [Looped Transformers as Programmable Computers](https://mlsanity.com/paper/2301.13196/) (2023)
- [Scaling Latent Reasoning via Looped Language Models](https://mlsanity.com/paper/2510.25741/) (2025)
- [Looped Transformers are Better at Learning Learning Algorithms](https://mlsanity.com/paper/2311.12424/) (2023)
- [Consciousness in Artificial Intelligence: Insights from the Science of Consciousness](https://mlsanity.com/paper/2308.08708/) (2023)
- [Verbalizable Representations Form a Global Workspace in Language Models](https://mlsanity.com/paper/2607.15495/) (2026)
- [A Mechanistic Analysis of Looped Reasoning Language Models](https://mlsanity.com/paper/2604.11791/) (2026)
- [Latent Chain-of-Thought? Decoding the Depth-Recurrent Transformer](https://mlsanity.com/paper/2507.02199/) (2025)

## Built on by (most cited first, 6 of 6)

- [Improving Test-Time Scaling with Adaptive Looped Transformers](https://mlsanity.com/paper/2609.35748/) (2026)
- [Residual Streams Read, Recurrent States Remember: The Global Workspace in Mamba Models](https://mlsanity.com/paper/2609.32102/) (2026)
- [Transformers Stop Thinking Too Early, and a Tiny LoRA Fixes It](https://mlsanity.com/paper/2609.36585/) (2026)
- [Rotate the Loop: Blockwise Conformality for Looped Transformers](https://mlsanity.com/paper/2610.14030/) (2026)
- [Monitoring latent reasoning in looped language models](https://mlsanity.com/paper/2610.18127/) (2026)
- [Learning to Compute Is Not Learning to Retain or Reuse in Looped Transformers](https://mlsanity.com/paper/2610.19768/) (2026)

## Abstract (arXiv)

Recent work identifies a mid-depth band of verbalisable, causally potent representations in a standard feedforward transformer --- a functional analogue of a global workspace. Whether the same workspace functionality emerges when depth is implemented through recurrence rather than a stack of distinct layers remains unknown. Looped and depth-recurrent transformers provide a direct test of this question because they reuse the same weights across depth. We extend the Jacobian lens to iterated architectures using a virtual-unrolling adapter. We apply the full workspace suite --- lens fitting, readout, and eleven causal experiment families --- to Ouro-2.6B (48 layers looped 4 times, deeply supervised) and Huginn-0125 (a 4-layer core recurred 16 times, trained for latent reasoning), using Qwen3.6-27B (64 untied layers) as the standard baseline. We find that a workspace forms in the iterated part of each architecture, but that recurrence changes how it can be accessed. Ouro reconstructs workspace content in every loop, and linear transport cannot carry that content across loop boundaries; writes and ablations must therefore span every remaining loop. Huginn carries content forward across all sixteen recurrences, while reads, writes, and ablations act only within a sliding window of roughly two recurrences. Whether newly injected content can be verbalised tracks explicit per-iteration supervision; whether existing content can be steered does not.