相关论文: A Mechanistic Account of Attention Sinks in GPT-2:…

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…

Attention sinks are tokens, often the beginning-of-sequence (BOS) token, that receive disproportionately high attention despite limited semantic relevance. In this work, we identify a class of attention sinks, which we term secondary sinks,…

Large Language Models (LLMs) tend to attend heavily to the first token in the sequence -- creating a so-called attention sink. Many works have studied this phenomenon in detail, proposing various ways to either leverage or alleviate it.…

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…

Large Language Models (LLMs) often assign disproportionate attention to the first token, a phenomenon known as the attention sink. Several recent approaches aim to address this issue, including Sink Attention in GPT-OSS and Gated Attention…

Despite the prevalence of the attention sink phenomenon in Large Language Models (LLMs), where initial tokens disproportionately monopolize attention scores, its structural origins remain elusive. This work provides a \textit{mechanistic…

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.…

Attention sink (AS) is a consistent pattern in transformer attention maps where certain tokens (often special tokens or positional anchors) disproportionately attract attention from other tokens. We show that in transformers, AS is not an…

Transformers often display an attention sink: probability mass concentrates on a fixed, content-agnostic position. Are sinks a byproduct of the optimization/training regime? Or are they sometimes functionally necessary in softmax…

Transformers underpin modern large language models (LLMs) and are commonly assumed to be behaviorally unstructured at random initialization, with all meaningful preferences emerging only through large-scale training. We challenge this…

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…

Recent studies have revealed various manifestations of position bias in transformer architectures, from the "lost-in-the-middle" phenomenon to attention sinks, yet a comprehensive theoretical understanding of how attention masks and…

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…

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…

Transformer models systematically favor certain token positions, yet the architectural origins of this position bias remain poorly understood. This bias is closely connected to the Lost-in-the-Middle phenomenon, where models underutilize…

The goal of this paper is to strengthen the reasoning of Omnimodal Large Language Models (Omni-LLMs) at inference time, without additional training. These models jointly process video, audio, and text, and given the large number of tokens…

Large Language Models (LLMs), despite their impressive capabilities, often fail to accurately repeat a single word when prompted to, and instead output unrelated text. This unexplained failure mode represents a vulnerability, allowing even…

Attention sinks are defined as tokens that attract disproportionate attention. While these have been studied in single modality transformers, their cross-modal impact in Large Vision-Language Models (LVLM) remains largely unexplored: are…

The Transformer is a fully attention-based alternative to recurrent networks that has achieved state-of-the-art results across a range of NLP tasks. In this paper, we analyze the structure of attention in a Transformer language model, the…

Attention-based architectures have become ubiquitous in machine learning, yet our understanding of the reasons for their effectiveness remains limited. This work proposes a new way to understand self-attention networks: we show that their…