Pith. sign in

REVIEW 6 cited by

Theory, Analysis, and Best Practices for Sigmoid Self-Attention

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2409.04431 v2 pith:I6TQADLX submitted 2024-09-06 cs.LG

classification cs.LG
keywords attentionsigmoidsoftmaxanalysistransformersworkattemptsbest
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Attention is a key part of the transformer architecture. It is a sequence-to-sequence mapping that transforms each sequence element into a weighted sum of values. The weights are typically obtained as the softmax of dot products between keys and queries. Recent work has explored alternatives to softmax attention in transformers, such as ReLU and sigmoid activations. In this work, we revisit sigmoid attention and conduct an in-depth theoretical and empirical analysis. Theoretically, we prove that transformers with sigmoid attention are universal function approximators and benefit from improved regularity compared to softmax attention. Through detailed empirical analysis, we identify stabilization of large initial attention norms during the early stages of training as a crucial factor for the successful training of models with sigmoid attention, outperforming prior attempts. We also introduce FLASHSIGMOID, a hardware-aware and memory-efficient implementation of sigmoid attention yielding a 17% inference kernel speed-up over FLASHATTENTION2 on H100 GPUs. Experiments across language, vision, and speech show that properly normalized sigmoid attention matches the strong performance of softmax attention on a wide range of domains and scales, which previous attempts at sigmoid attention were unable to fully achieve. Our work unifies prior art and establishes best practices for sigmoid attention as a drop-in softmax replacement in transformers.

Discussion (0). Sign in to comment.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Complex-Valued Phase-Coherent Transformer

    cs.LG 2026-05 unverdicted novelty 7.0 of 10 partial

    Sigmoid gating on L2-normalised complex cosine scores, with no row normalisation, generalises across long-range, positional, phase and vision tasks, though the depth-stability theorem assumes its own substance.

  2. Legible-by-Construction: Attention and End-to-End Transformers

    cs.CL 2026-07 conditional novelty 6.0 of 10

    Bounding attention values into fuzzy memberships or Boolean set operations yields 44–62% crisp selective detectors and an end-to-end legible transformer at baseline quality.

  3. Unpacking Softmax: How Temperature Drives Representation Collapse, Compression, and Generalization

    cs.LG 2025-06 conditional novelty 6.0 of 10

    Softmax temperature controls the rank of learned representations: high temperature induces rank-deficit bias, compresses features, hurts OOD generalization, and improves OOD detection.

  4. Your Attention Matters: to Improve Model Robustness to Noise and Spurious Correlations

    cs.LG 2025-07 conditional novelty 5.0 of 10

    Doubly stochastic attention is the most robust of five ViT attention mechanisms to fog corruption in relative accuracy, based on single-seed experiments on CIFAR-10, CIFAR-100, and Imagenette.

  5. A Theoretical Study of (Hyper) Self-Attention through the Lens of Interactions: Representation, Training, Generalization

    cs.LG 2025-06 conditional novelty 5.0 of 10

    Single-layer linear self-attention can represent, train on, and length-generalize pairwise interaction functions under data-versatility and exact-realizability assumptions, and the paper introduces higher-order HyperA...

  6. Scaling Context Requires Rethinking Attention

    cs.LG 2025-07 conditional novelty 4.0 of 10

    On 64k-token natural language training, power attention with degree 2 achieves lower loss per FLOP than both softmax attention and existing linear attention, and its GPU kernels run faster than Flash Attention at long...

Pith tools