arXiv AI
Aug 11
Full-bandwidth transformer
arXiv:2608. 08888v1 Announce Type: new Abstract: Autoregressive transformers compute along two axes: horizontally across generated tokens, and vertically through model depth.
SATFormer: Learning When to Use Early Value Residuals in Transformers
Related stories
arXiv AI
Jul 24
Scaling Interpretable Transformers with Parity Bottleneck Layers
arXiv:2607. 20652v1 Announce Type: cross Abstract: Language models are thought to exhibit the phenomenon of superposition, representing many more features than dimensions in their residual streams.
arXiv Machine Learning
Jun 29
Prism Transformer: Progressive Head Schedules for Hierarchical Attention Processing
arXiv:2606. 27449v1 Announce Type: new Abstract: Multi-head attention conventionally partitions the hidden dimension equally across all heads at every layer, enforcing an identical representational subspace dimension (dh = dmodel/h) throughout the models depth.
arXiv AI
Jun 29
The Context-Ready Transformer
arXiv:2606. 27538v1 Announce Type: cross Abstract: We introduce the context-ready transformer, a new recurrent neural network architecture built from a D-layer transformer block that pre-contextualizes each token before it enters the block.
arXiv AI
Jun 26
CascadeFormer: Depth-Tapered Transformers Motivated by Gradient Fan-in Asymmetry
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 Computation and Language
Aug 25
Do Value Vectors in Deep Layers Need Context from the Residual Stream?
The paper investigates whether deep transformer layers require context from the residual stream to compute value vectors. It finds that allowing deeper layers to use a context‑free value vector—preserving original token information—significantly improves performance, and adding context afterward yields little extra benefit. The authors introduce Bank of Values (BoV), a lookup table of token‑specific value vectors for the last third of layers, which reduces compute and memory while matching or surpassing prior methods on large models.
arXiv Computation and Language
Oct 21
Asymmetric Looped Transformers: Fixed Global Representations for an Iterative Decoder
The Asymmetric Looped Transformer (AsymLoop) separates context encoding from state refinement by using a single-pass causal encoder to produce fixed global key-value caches and a weight-tied recurrent decoder to generate local key-values from its evolving state. Joint attention over both sources allows each update to combine persistent contextual information with the current intermediate state. Experiments on FineWeb-Edu across models from 60M to 1B show lower validation perplexity and higher downstream accuracy than full-loop models, with a 1B model improving zero-shot commonsense reasoning accuracy from 49.58% to 50.85%.
arXiv AI
Jun 16
Scaling Adaptive Depth with Norm-Agnostic Residual Networks
arXiv:2606. 16112v1 Announce Type: cross Abstract: Residual architectures are ubiquitous in deep learning, but they suffer from a subtle structural limitation: the norm of the residual stream can grow rapidly with depth.
arXiv Machine Learning
Aug 26
Gated Recurrent Transformers: Expressive Depth through Recurrent Modulation in Transformers
arXiv:2608.15062v3 Announce Type: replace-cross Abstract: Scaling transformer language models creates an inherent tension between expressivity and memory efficiency. While unique weights across layer...
arXiv Machine Learning
Aug 27
Gated Recurrent Transformers: Expressive Depth through Recurrent Modulation
The paper introduces Gated Recurrent Transformers, a depth‑sharing architecture that brackets a single shared core with fixed prelude and coda blocks and uses a lightweight projection and element‑wise update gate to modulate recurrent updates. This design allows functional specialization across recurrences while reducing memory footprint. Experiments show that, under equal FLOPs or parameter budgets, the recurrent model matches or surpasses deeper GPT‑2 Small baselines, achieving similar or better accuracy with fewer parameters and lower peak decoding memory.
arXiv Machine Learning
Sep 30
Procedural Core: A Compact Recurrent Initialization for Vision Transformers
arXiv:2609.37631v1 Announce Type: new Abstract: Transformers are typically trained from random initialization, requiring all their capabilities to emerge from large-scale optimization. Recent work sh...
arXiv Computer Vision
Sep 14
DRDN: Decoupled Representation Dynamic Network for From-Scratch ViT Class-Incremental Learning
arXiv:2607.01630v2 Announce Type: replace Abstract: Dynamic expansion methods for class-incremental learning (CIL) protect task-specific knowledge by growing dedicated tokens or subnetworks, yet our...