LT2: Linear-Time Looped Transformers

• URL: http://arxiv.org/abs/2605.20670v2
• Date: Fri, 22 May 2026 21:41:56 GMT
• Title: LT2: Linear-Time Looped Transformers
• Abstract summary: We introduce LT2, a family of looped architectures that replace quadratic softmax attention with subquadratic, linear-time attention.<n>We show consistent empirical gains across controlled recall, state-tracking, and language modeling tasks.<n>Two variants are especially promising: LT2-hybrid (GDN+DSA), which interleaves linear and sparse attention to maximize efficiency and matches the standard looped transformer's quality at fully linear-time cost; and LT2-hybrid (Full+GDN), which interleaves GDN with a small fraction of full attention layers to maximize quality, surpassing the standard looped transformer in both
• Score: 24.49487079056893
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Looped Transformers (LT) have emerged as a powerful architecture by iterating their layers multiple times before decoding the final token. However, pairing them with full attention retains quadratic complexity, making them computationally expensive and slow. We introduce LT2 (Linear-Time Looped Transformers), a family of looped architectures that replace quadratic softmax attention with subquadratic, linear-time attention. We study two variants: LT2-linear with linear attention and LT2-sparse with sparse attention. We find that looping uniquely synergizes with these variants: it enables iterative memory refinement in linear attention and progressively expands the effective receptive field in sparse attention. We formalize these benefits theoretically and demonstrate consistent empirical gains across controlled recall, state-tracking, and language modeling tasks. We then explore LT2-hybrid, which combines different attention variants in a looped setting. Two variants are especially promising: LT2-hybrid (GDN+DSA), which interleaves linear and sparse attention to maximize efficiency and matches the standard looped transformer's quality at fully linear-time cost; and LT2-hybrid (Full+GDN), which interleaves GDN with a small fraction of full attention layers to maximize quality, surpassing the standard looped transformer in both performance and efficiency. We also show how to convert a pre-trained LT into an LT2-hybrid model. With about 1B tokens of training, our converted model, Ouro-hybrid-1.4B, outperforms industry-level 1B models and is competitive with industry-level 4B models while retaining the speed benefits of linear-time attention. Together, these results show a clear path toward making looped transformers more scalable and advancing efficient, capable small language models.

Related papers

• SMELT: Scaling Laws for Compute-Matched MoE Looped Transformers [50.00050394179417]
  Looped Transformers increase effective depth by iterating a shared block of layers, but most evaluations compare at fixed model size.<n>We study looping on Mixture-of-Experts Transformers while closely matching per-token FLOPs, total non-embedding parameters, and KV cache.<n>Through a series of ablations, we arrive at a recipe we call SMELT, which loops the middle half of layers twice while matching the unlooped Baseline on all three budgets.
  arXiv  Detail & Related papers  (2026-09-01T14:52:57Z)
• LINA: Linear Autoregressive Image Generative Models with Continuous Tokens [56.80443965097921]
  Autoregressive models with continuous tokens form a promising paradigm for visual generation, especially for text-to-image (T2I) synthesis.<n>We study how to design compute-efficient linear attention within this framework.<n>We present LINA, a simple and compute-efficient T2I model built entirely on linear attention, capable of generating high-fidelity 1024x1024 images from user instructions.
  arXiv  Detail & Related papers  (2026-01-30T06:44:33Z)
• Parallel Loop Transformer for Efficient Test-Time Computation Scaling [34.8255648599522]
  Large Language Models (LLMs) are powerful but often too slow and costly for real-world use during inference.<n>Looped transformers save on parameters by reusing the same weights for multiple computational steps, or "loops"<n>This approach has a major flaw: the loops run one after another, causing inference latency and memory requirements to increase with each added loop.
  arXiv  Detail & Related papers  (2025-10-28T15:35:50Z)
• Systems and Algorithms for Convolutional Multi-Hybrid Language Models at Scale [68.6602625868888]
  We introduce convolutional multi-hybrid architectures, with a design grounded on two simple observations.<n>Operators in hybrid models can be tailored to token manipulation tasks such as in-context recall, multi-token recall, and compression.<n>We train end-to-end 1.2 to 2.9 times faster than optimized Transformers, and 1.1 to 1.4 times faster than previous generation hybrids.
  arXiv  Detail & Related papers  (2025-02-25T19:47:20Z)
• The Mamba in the Llama: Distilling and Accelerating Hybrid Models [76.64055251296548]
  We show how to distill large Transformers into linear RNNs by reusing the linear projection weights from attention layers with academic GPU resources.<n>The resulting hybrid model achieves performance comparable to the original Transformer in chat benchmarks.<n>We also introduce a hardware-aware speculative decoding algorithm that accelerates the inference speed of Mamba and hybrid models.
  arXiv  Detail & Related papers  (2024-08-27T17:56:11Z)
• Parallelizing Linear Transformers with the Delta Rule over Sequence Length [49.88826673324244]
  This work describes a hardware-efficient algorithm for training linear transformers with the delta rule.<n>We train a 1.3B model for 100B tokens and find that it outperforms recent linear-time baselines.
  arXiv  Detail & Related papers  (2024-06-10T17:24:42Z)
• Gated Linear Attention Transformers with Hardware-Efficient Training [60.670102007737476]
  This work describes a hardware-efficient algorithm for linear attention that trades off memory movement against parallelizability. We then generalize this algorithm to a more expressive variant of linear attention with data-dependent gates. When used as a replacement for the standard attention layer in Transformers, the resulting gated linear attention Transformer is found to perform competitively.
  arXiv  Detail & Related papers  (2023-12-11T18:51:59Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.