Looped Transformers increase effective depth by repeatedly applying shared layers, offering a way to scale computation without increasing parameter count. However, full-stack looping places context construction and iterative state refinement within the same recurrent computation, without explicitly distinguishing their roles. We introduce the Asymmetric Looped Transformer (AsymLoop), which separates single-pass context encoding from recurrent state refinement. A single-pass causal encoder produces layer-specific Global key-value (KV) caches that remain fixed across iterations, while a weight-tied recurrent decoder generates Local KV from its evolving state. Joint attention over both sources under a single softmax allows each update to integrate persistent contextual information with the current intermediate state. For training, we apply a heavy-tail-guided layerwise learning-rate method to the encoder and decoder. Experiments on FineWeb-Edu across four model scales, from 60M to 1B, show consistently lower validation perplexity and higher average downstream accuracy than Full Loop under matched effective depth and training token budgets. At 1B, AsymLoop improves average zero-shot accuracy across seven commonsense reasoning benchmarks from 49.58% to 50.85%. Further experiments extend these benefits to longer training budgets, an OLMo-style backbone, and sparse mixture-of-experts models. These results support asymmetric context encoding and state refinement as an effective design for recurrent language models.
This work proposes LoopSpec, a training-free self-speculative decoding framework tailored for Looped Transformers that introduces a selective second proposal from deeper recurrent depth while ensuring lossless decoding under both greedy and sampling regimes.
SangLyul Cho, Lang-Qing Cui, Sehoon Kim et al.· arXiv.org· 7 citations· ⚡1
Looped Transformers increase computation per token by repeatedly applying a shared backbone, offering a way to improve language modeling and downstream task performance at a fixed backbone parameter count. However, each loop depends on the preceding loop, so increasing recurrent depth also increases decoding latency. E...
Fei Song, Bo-Yi Zeng, Yi-Qing Hao et al.· 0 citations
Looped Transformers achieve parameter efficiency by repeatedly executing a shared block across recurrent loops. Each loop yields an intermediate representation decodable for the same next token, yet standard decoding discards earlier states. Because earlier loops embody less computation, recurrence inherently supplies...
Weihao Liu, Huangjie Zheng, Tian-Rong Chen et al.· 2 citations
This framework derives loop transitions in closed form from a projection, a local objective and an optimizer update rule, which combines explicit weight decay, adaptive step size and a delta objective and improves average generative performance over the compared looped and non-looped baselines under matched training FL...
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...
Xuan Ouyang, Ze-Fan Cai, Chen-Long Zhang et al.· 0 citations
Looped transformers increase effective depth by repeatedly applying a shared set of transformer layers, offering a parameter-efficient way to scale computation. However, standard residual connections aggregate information across recurrent depth into a single stream, providing limited control over which intermediate rep...
Yue-Kun Yao, Yu-Pei Du, Da-Wei Zhu et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.