arXiv Machine Learning
Sep 22
Complex-valued Phase-Coherent Transformers
arXiv:2609. 22415v1 Announce Type: new Abstract: Complex-valued Transformers have inherited softmax attention over the raw complex inner product.
Complex-Valued Phase-Coherent Transformer
The paper introduces the Phase-Coherent Transformer (PCT), a complex-valued architecture that replaces traditional softmax attention with a real-valued, smooth gate applied to L2-normalised query-key similarities. PCT eliminates token competition, preserving phase information across layers, and demonstrates strong generalisation on a variety of mid-scale benchmarks, outperforming both standard softmax Transformers and other complex-valued counterparts. Experiments confirm that the gate design is essential: preserving negatively aligned phase components is crucial for performance, while violating these conditions leads to degradation or collapse on long-range tasks.
Related stories
arXiv AI
Jul 17
Language as a Wave Phenomenon: Semantic Phase Locking and Interference in Neural Networks
arXiv:2512. 01208v5 Announce Type: replace-cross Abstract: In standard Transformer architectures, semantic importance is often conflated with activation magnitude, obscuring the geometric structure of latent representations.
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 Machine Learning
Aug 10
The Expressive Power of Low Precision Softmax Transformers with (Summarized) Chain-of-Thought
arXiv:2605. 18079v2 Announce Type: replace Abstract: Existing expressivity results for transformers typically rely on hardmax attention, high precision, and other architectural modifications that disconnect them from the models used in practice.
arXiv Machine Learning
Sep 1
Tracing distinguishability through transformer processing with stochastic LayerNorm
arXiv:2608.30720v1 Announce Type: new Abstract: Representational similarity is foundational to analyses of deep networks, yet distances between point-valued representations are not intrinsically tied...
arXiv Machine Learning
Jun 5
Exact Linear Attention
arXiv:2605. 18848v3 Announce Type: replace Abstract: This paper introduces Exact Linear Attention (ELA), a mechanism that achieves linear computational complexity for Transformer attention by exploiting the exact decomposition property of kernel functions, thereby eliminating approximation error.
arXiv Machine Learning
Sep 2
Performance-Efficiency Tradeoffs in Transformers: An Approximation Theory Perspective
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.
arXiv AI
Jun 30
Emergence of Minimal Circuits for Indirect Object Identification in Attention-Only Transformers
arXiv:2510. 25013v2 Announce Type: replace-cross Abstract: Mechanistic interpretability aims to reverse-engineer large language models (LLMs) into human-understandable computational circuits.
arXiv AI
Jul 10
The Phasor Transformer: Resolving Attention Bottlenecks on the Unit Circle
arXiv:2603. 17433v2 Announce Type: replace-cross Abstract: Transformer models have redefined sequence learning, yet dot-product self-attention introduces a quadratic token-mixing bottleneck for long-context time-series.
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
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.