English
Related papers

Related papers: Attention Sinks: A 'Catch, Tag, Release' Mechanism…

200 papers

The KV-Cache technique has become the standard for the inference of large language models (LLMs). Yet, it is widely criticized that KV-Cache can become a bottleneck of the LLM inference system. This paper enables a novel dynamic KV-Cache…

Computation and Language · Computer Science 2025-04-18 Zihao Zeng , Bokai Lin , Tianqi Hou , Hao Zhang , Zhijie Deng

Auto-regressive sequence-to-sequence models with attention mechanism have achieved state-of-the-art performance in many tasks such as machine translation and speech synthesis. These models can be difficult to train. The standard approach,…

Machine Learning · Computer Science 2019-10-04 Qingyun Dou , Yiting Lu , Joshua Efiong , Mark J. F. Gales

Efficient long-context inference is critical as large language models (LLMs) adopt context windows of ranging from 128K to 1M tokens. However, the growing key-value (KV) cache and the high computational complexity of attention create…

Computation and Language · Computer Science 2025-03-13 Guangtao Wang , Shubhangi Upasani , Chen Wu , Darshan Gandhi , Jonathan Li , Changran Hu , Bo Li , Urmish Thakker

In this paper we explore where information is collected and how it is propagated throughout layers in large language models (LLMs). We begin by examining the surprising computational importance of punctuation tokens which previous work has…

Computation and Language · Computer Science 2025-08-21 Sonakshi Chauhan , Maheep Chaudhary , Koby Choy , Samuel Nellessen , Nandi Schoots

Stance detection on social media is challenging for Large Language Models (LLMs), as emerging slang and colloquial language in online conversations often contain deeply implicit stance labels. Chain-of-Thought (COT) prompting has recently…

Computation and Language · Computer Science 2023-10-31 Joseph Gatto , Omar Sharif , Sarah Masud Preum

The self-attention mechanism prevails in modern machine learning. It has an interesting functionality of adaptively selecting tokens from an input sequence by modulating the degree of attention localization, which many researchers speculate…

Machine Learning · Statistics 2024-02-06 Han Bao , Ryuichiro Hataya , Ryo Karakida

Sentence encoders map sentences to real valued vectors for use in downstream applications. To peek into these representations - e.g., to increase interpretability of their results - probing tasks have been designed which query them for…

Computation and Language · Computer Science 2020-10-29 Steffen Eger , Johannes Daxenberger , Iryna Gurevych

Large language models (LLMs) are increasingly deployed on complex reasoning tasks, yet little is known about their ability to internally evaluate problem difficulty, which is an essential capability for adaptive reasoning and efficient…

Computation and Language · Computer Science 2025-10-14 Sunbowen Lee , Qingyu Yin , Chak Tou Leong , Jialiang Zhang , Yicheng Gong , Shiwen Ni , Min Yang , Xiaoyu Shen

While large language models (LLMs) demonstrate remarkable success in multilingual translation, their internal core translation mechanisms, even at the fundamental word level, remain insufficiently understood. To address this critical gap,…

Computation and Language · Computer Science 2026-01-16 Hongbin Zhang , Kehai Chen , Xuefeng Bai , Xiucheng Li , Yang Xiang , Min Zhang

Harmful fine-tuning can invalidate safety alignment of large language models, exposing significant safety risks. In this paper, we utilize the attention sink mechanism to mitigate harmful fine-tuning. Specifically, we first measure a…

Artificial Intelligence · Computer Science 2026-02-12 Guozhi Liu , Weiwei Lin , Tiansheng Huang , Ruichao Mo , Qi Mu , Xiumin Wang , Li Shen

Masked diffusion models (MDMs), which leverage bidirectional attention and a denoising process, are narrowing the performance gap with autoregressive models (ARMs). However, their internal attention mechanisms remain under-explored. This…

Machine Learning · Computer Science 2026-01-14 Xin Dai , Pengcheng Huang , Zhenghao Liu , Shuo Wang , Yukun Yan , Chaojun Xiao , Yu Gu , Ge Yu , Maosong Sun

In-context learning (ICL) allows large models to adapt to tasks using a few examples, yet its extension to vision-language models (VLMs) remains fragile. Our analysis reveals that the fundamental limitation lies in an inductive gap, models…

Computer Vision and Pattern Recognition · Computer Science 2026-05-05 Haoyu Wang , Haonan Wang , Yuyan Chen , Jun Chen , Gang Liu , Qian Wang , Jiahong Yan , Yanghua Xiao

Language models are often trained to maximize the likelihood of the next token given past tokens in the training dataset. However, during inference time, they are utilized differently, generating text sequentially and auto-regressively by…

Machine Learning · Computer Science 2025-01-22 Zhepeng Cen , Yao Liu , Siliang Zeng , Pratik Chaudhari , Huzefa Rangwala , George Karypis , Rasool Fakoor

Context compression is an advanced technique that accelerates large language model (LLM) inference by converting long inputs into compact representations. Existing methods primarily rely on autoencoding tasks to train special compression…

Computation and Language · Computer Science 2026-03-12 Xin Liu , Runsong Zhao , Pengcheng Huang , Xinyu Liu , Junyi Xiao , Chunyang Xiao , Tong Xiao , Shengxiang Gao , Zhengtao Yu , Jingbo Zhu

We consider the problem of predicting edges in a graph from node attributes in an e-commerce setting. Specifically, given nodes labelled with search query text, we want to predict links to related queries that share products. Experiments…

Machine Learning · Computer Science 2020-06-15 Matthew Dippel , Adam Kiezun , Tanay Mehta , Ravi Sundaram , Srikanth Thirumalai , Akshar Varma

Many empirical studies have provided evidence for the emergence of algorithmic mechanisms (abilities) in the learning of language models, that lead to qualitative improvements of the model capabilities. Yet, a theoretical characterization…

Machine Learning · Computer Science 2025-02-10 Hugo Cui , Freya Behrens , Florent Krzakala , Lenka Zdeborová

We study how large language models (LLMs) reason about memorized knowledge through simple binary relations such as equality ($=$), inequality ($<$), and inclusion ($\subset$). Unlike in-context reasoning, the axioms (e.g., $a < b, b < c$)…

Machine Learning · Computer Science 2025-09-18 Jonathan Shaki , Emanuele La Malfa , Michael Wooldridge , Sarit Kraus

Tagging systems play an essential role in various information retrieval applications such as search engines and recommender systems. Recently, Large Language Models (LLMs) have been applied in tagging systems due to their extensive world…

Information Retrieval · Computer Science 2025-05-07 Ruiming Tang , Chenxu Zhu , Bo Chen , Weipeng Zhang , Menghui Zhu , Xinyi Dai , Huifeng Guo

Prior work on in-context copying has shown the existence of induction heads, which attend to and promote individual tokens during copying. In this work we discover a new type of induction head: concept-level induction heads, which copy…

Computation and Language · Computer Science 2025-07-22 Sheridan Feucht , Eric Todd , Byron Wallace , David Bau

Humans can approach complex visual problems by mentally simulating intermediate visual steps, rather than reasoning through language alone. Inspired by this, several works on Vision-Language Models have recently explored chain-of-thought…

Computer Vision and Pattern Recognition · Computer Science 2026-05-20 André G. Viveiros , Nuno Gonçalves , André F. T. Martins , Matthias Lindemann