Related papers: Attention Sinks Are Provably Necessary in Softmax …

Language Models (LMs) assign significant attention to the first token, even if it is not semantically important, which is known as attention sink. This phenomenon has been widely adopted in applications such as streaming/long context…

Attention architectures are widely used; they recently gained renewed popularity with Transformers yielding a streak of state of the art results. Yet, the geometrical implications of softmax-attention remain largely unexplored. In this work…

The Transformer architecture, a cornerstone of modern Large Language Models (LLMs), has achieved extraordinary success in sequence modeling, primarily due to its attention mechanism. However, despite its power, the standard attention…

Transformer attention is typically implemented using softmax normalization, which enforces attention weights with unit sum normalization. While effective in many settings, this constraint can limit flexibility in controlling attention…

Attention sinks -- tokens that receive disproportionate attention mass -- are assumed to be functionally important in autoregressive language models, but their role in diffusion transformers remains unclear. We present a causal analysis in…

We introduce softpick, a rectified, not sum-to-one, drop-in replacement for softmax in transformer attention mechanisms that eliminates attention sink and massive activations. Our experiments with 340M and 1.8B parameter models demonstrate…

Transformers commonly exhibit an attention sink: disproportionately high attention to the first position. We study this behavior in GPT-2-style models with learned query biases and absolute positional embeddings. Combining structural…

We investigate the functional role of emergent outliers in large language models, specifically attention sinks (a few tokens that consistently receive large attention logits) and residual sinks (a few fixed dimensions with persistently…

This paper questions whether the strong performance of softmax attention in transformers stems from producing a probability distribution over inputs. Instead, we argue that softmax's effectiveness lies in its implicit regularization of the…

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…

Attention based models such as Transformers involve pairwise interactions between data points, modeled with a learnable attention matrix. Importantly, this attention matrix is normalized with the SoftMax operator, which makes it row-wise…

Attention sinks and massive activations are recurring and closely related phenomena in Transformer models. Existing explanations have largely focused on the forward pass, yet in pre-norm Transformers, large residual-stream norms play only…

The softmax function is crucial in Transformer attention, which normalizes each row of the attention scores with summation to one, achieving superior performances over other alternative functions. However, the softmax function can face a…

Self-attention in Transformers relies on globally normalized softmax weights, causing all tokens to compete for influence at every layer. When composed across depth, this interaction pattern induces strong synchronization dynamics that…

Large language models (LLMs) often concentrate their attention on a few specific tokens referred to as attention sinks. Common examples include the first token, a prompt-independent sink, and punctuation tokens, which are prompt-dependent.…

We study two recurring phenomena in Transformer language models: massive activations, in which a small number of tokens exhibit extreme outliers in a few channels, and attention sinks, in which certain tokens attract disproportionate…

Large language models rely on attention mechanisms with a softmax activation. Yet the dominance of softmax over alternatives (e.g., component-wise or linear) remains poorly understood, and many theoretical works have focused on the…

Large Language Models (LLMs) often allocate disproportionate attention to specific tokens, a phenomenon commonly referred to as the attention sink. While such sinks are generally considered detrimental, prior studies have identified a…

As the foundational architecture of modern machine learning, Transformers have driven remarkable progress across diverse AI domains. Despite their transformative impact, a persistent challenge across various Transformers is Attention Sink…

Practitioners have consistently observed three puzzling phenomena in transformer-based large language models (LLMs): attention sinks, value-state drains, and residual-state peaks, collectively referred to as extreme-token phenomena. These…