arXiv:2606. 18524v1 Announce Type: new Abstract: Looped (weight-tied) Transformers apply a shared residual block $N$ times ($h \leftarrow h + \varepsilon\,f(h)$, same $f$ at each step), increasing effective depth without adding parameters.
arXiv:2606. 26538v1 Announce Type: cross Abstract: Deep Transformers are composed of uniformly stacked residual blocks, yet their deepest layers often add little value.
arXiv:2607. 13491v1 Announce Type: cross Abstract: Looped Transformers scale sequential computation by applying a compact stack of physical blocks for multiple rounds, increasing unrolled depth without increasing stored parameters.
arXiv:2605. 09165v2 Announce Type: replace Abstract: Looped language models repeat a set of transformer layers through depth, reducing memory costs and providing natural early-exit points at loop boundaries.
arXiv:2606. 31859v1 Announce Type: new Abstract: Residual connections add every sublayer's proposed update with a fixed coefficient of one; the network never evaluates whether an update is reliable before committing it.
arXiv:2607. 10681v1 Announce Type: new Abstract: In pre-LayerNorm looped transformers, LayerNorm inside the recurrent block acts as an implicit gain controller: by coupling the block's local Lipschitz constant inversely to the activation scale, it renders the recurrence Jacobian non-normal -- asymptotically contractive at every verified fixed point even where its operator norm exceeds 1 -- so the true stability budget is the spectral margin, not an operator-norm bound.
arXiv:2510. 09904v2 Announce Type: replace-cross Abstract: Despite their widespread use, training deep Transformers can be unstable.
The paper demonstrates that architectural changes—specifically looped transformers and boundary operators—can alter scaling exponents in pre‑training, yielding exponential performance gains for a given computational budget. Looping, or recursive depth, enables model growth that matches larger models (e.g., a 7.4B looped architecture matching GPT‑3 13B) with significantly less compute, while boundary operators provide additional, though smaller, efficiency improvements. In data‑constrained, multi‑epoch scenarios, increasing loops with scale serves as a useful regularizer, suggesting that deeper computational depth drives compute‑efficiency gains that grow with model size.
arXiv:2605. 20708v2 Announce Type: replace-cross Abstract: Diffusion Transformers (DiTs) have become a de facto backbone of modern visual generation, and nearly every major axis of their design -- tokenization, attention, conditioning, objectives, and latent autoencoders -- has been extensively revisited.
The paper introduces LOOM, a method for scaling looped mixture‑of‑experts (MoE) Transformers beyond the typical two‑loop limit. LOOM addresses two key obstacles: it stabilizes deep recurrence by bounding residual variance and re‑injecting the input embedding, and it prevents expert selection collapse by using per‑loop routers and a looping residual to maintain computational diversity. Experiments on 100 M–1.7 B parameter models show stable scaling to 9–12 loops, with significant perplexity reductions and zero‑shot accuracy gains under near‑iso‑FLOP conditions.
We investigate how each component of the Transformer feedforward block architecture design determines how much rank survives across depth at initialization. We reinterpret skip connections and normalization, long understood as controlling magnitude, as mechanisms for preserving gradient rank across depth, since the very matrix multiplications and nonlinear activations that make the network expressive also reduce the rank.
The paper introduces the Drift Contract, a spectral update geometry for local learning that improves depth robustness and hyperparameter stability. By applying momentum orthogonalization with spectral step scaling to per‑layer updates, the authors achieve consistent performance across a wide range of widths and depths on CIFAR‑10 MLPs, outperforming local Adam and providing a per‑layer, input‑conditioned drift bound. The study also shows that the spectral geometry itself, rather than step‑size rules, drives the observed depth robustness, while a negative result indicates that the stability benefit is limited to non‑normalized layers.
The paper examines how to allocate attention heads and head dimensions across Transformer layers to balance expressivity and efficiency. It provides a mathematical analysis of early layers’ role in information extraction and characterizes the trade‑off between head count and dimension under a fixed parameter budget. The authors prove a saturation effect of softmax activations, showing that increasing head dimensions yields diminishing returns, especially for long sequences, and propose strategies for efficient parameter allocation across layers.