Trendoday

Loopback Transformer: Towards a Unified View of Residual Looped Transformers

Looped Transformers reduce parameter count by repeatedly applying shared blocks, but parameter sharing alone does not determine how the results of repeated computation are reused. We take the depth-me

  • Upvotes11
–Rank now
#9Peak rank
2Days on list
225Score · 7d
Nov 16, 2026First seen
Rank · 7 days

About this paper

A looped Transformer that separately controls input strength, total update weight, and per-update allocation achieves lower validation loss and higher downstream accuracy than standard and looped residual baselines while using fewer parameters.

Looped Transformers reduce parameter count by repeatedly applying shared blocks, but parameter sharing alone does not determine how the results of repeated computation are reused. We take the depth-memory view of residual connections and analyze which representation is stored after each block application and how later reads weight it. A full block output contains both its input and its new update, so weighting that output also reweights the history already carried by the input. Reading individual updates instead, with the shared stack's input kept on a separate path, gives separate controls over input strength, total update weight, and allocation across updates. The Loopback Transformer…

Looped Transformersparameter sharingshared blocksresidual connectionsdepth-memory view of residual connectionsblock outputinput strengthtotal update weightallocation across updatesLoopback Transformer
Upvotes
11
Authors
6
Published
Nov 4, 2026

Summary and abstract excerpt via Hugging Face Papers ↗ · arXiv.

Trending history

Loopback Transformer: Towards a Unified View of Residual Looped Transformers was first seen on Hugging Face · Daily papers on Nov 16, 2026. Since then it has appeared in 13 checks on 2 different days. Its best position was #9 of 30, reached on Nov 17, 01:28 UTC. It dropped off the list after Nov 17, 05:28 UTC.

Rank history

Position on Hugging Face · Daily papers at every check (lower is better).

#1#11#20#30 05:2811:2817:2823:2805:28 Nov 16, 05:28 UTC · #19Nov 16, 07:28 UTC · #18Nov 16, 09:28 UTC · #15Nov 16, 11:28 UTC · #16Nov 16, 13:28 UTC · #19Nov 16, 15:28 UTC · #16Nov 16, 17:28 UTC · #16Nov 16, 19:28 UTC · #12Nov 16, 21:28 UTC · #12Nov 16, 23:28 UTC · #10Nov 17, 01:28 UTC · #9Nov 17, 03:28 UTC · #9Nov 17, 05:28 UTC · #9

Last 14 days

Best rank per day (UTC).

Nov 9today

Metric history

Platform metric as reported by Hugging Face.

071421 05:2811:2817:2823:2805:28 Nov 16, 05:28 UTC · 11Nov 16, 07:28 UTC · 13Nov 16, 09:28 UTC · 14Nov 16, 11:28 UTC · 14Nov 16, 13:28 UTC · 14Nov 16, 15:28 UTC · 15Nov 16, 17:28 UTC · 17Nov 16, 19:28 UTC · 18Nov 16, 21:28 UTC · 18Nov 16, 23:28 UTC · 19Nov 17, 01:28 UTC · 20Nov 17, 03:28 UTC · 21Nov 17, 05:28 UTC · 21

Questions

Is Loopback Transformer: Towards a Unified View of Residual Looped Transformers trending on HF Daily Papers right now?

Not at the moment. It was last on the list on Nov 17, 05:28 UTC, after 2 days of trending.

What is the highest rank Loopback Transformer: Towards a Unified View of Residual Looped Transformers reached?

#9 on Hugging Face · Daily papers, on Nov 17, 01:28 UTC. Trendoday checks this list every 120 minutes.

How long has Loopback Transformer: Towards a Unified View of Residual Looped Transformers been trending?

It was first seen on Nov 16, 2026 and has been on the list on 2 days (13 checks) since.