English
Related papers

Related papers: Task Vectors in In-Context Learning: Emergence, Fo…

200 papers

In-context learning (ICL) allows large language models (LLMs) to adapt to new tasks through demonstrations, yet it suffers from escalating inference costs as context length increases. While task vectors offer a promising alternative by…

Computation and Language · Computer Science 2026-05-21 Jihoon Kwon , Jiwon Choi , Jy-yong Sohn

Pre-trained text-to-text transformers such as BART have achieved impressive performance across a range of NLP tasks. Recent study further shows that they can learn to generalize to novel tasks, by including task descriptions as part of the…

Computation and Language · Computer Science 2021-06-16 Qinyuan Ye , Xiang Ren

Transformers have proven highly effective across various applications, especially in handling sequential data such as natural languages and time series. However, transformer models often lack clear interpretability, and the success of…

Machine Learning · Computer Science 2025-12-01 Wei Shi , Yuan Cao

Fully test-time adaptation aims to adapt the network model based on sequential analysis of input samples during the inference stage to address the cross-domain performance degradation problem of deep neural networks. This work is based on…

Computer Vision and Pattern Recognition · Computer Science 2024-07-18 Yushun Tang , Shuoshuo Chen , Zhehan Kan , Yi Zhang , Qinghai Guo , Zhihai He

Transformer models exhibit in-context learning: the ability to accurately predict the response to a novel query based on illustrative examples in the input sequence. In-context learning contrasts with traditional in-weights learning of…

Machine Learning · Computer Science 2023-12-07 Gautam Reddy

Pre-trained large language models have demonstrated a strong ability to learn from context, known as in-context learning (ICL). Despite a surge of recent applications that leverage such capabilities, it is by no means clear, at least…

Artificial Intelligence · Computer Science 2025-10-28 Bingqing Song , Jiaxiang Li , Rong Wang , Songtao Lu , Mingyi Hong

In-context learning (ICL) allows transformer-based language models that are pre-trained on general text to quickly learn a specific task with a few "task demonstrations" without updating their parameters, significantly boosting their…

Computation and Language · Computer Science 2024-12-17 Zijian Zhou , Xiaoqiang Lin , Xinyi Xu , Alok Prakash , Daniela Rus , Bryan Kian Hsiang Low

Context plays an important role in human language understanding, thus it may also be useful for machines learning vector representations of language. In this paper, we explore an asymmetric encoder-decoder structure for unsupervised…

Neural and Evolutionary Computing · Computer Science 2018-06-04 Shuai Tang , Hailin Jin , Chen Fang , Zhaowen Wang , Virginia R. de Sa

Transformers are deep architectures that define "in-context mappings" which enable predicting new tokens based on a given set of tokens (such as a prompt in NLP applications or a set of patches for a vision transformer). In this work, we…

Computation and Language · Computer Science 2024-10-04 Takashi Furuya , Maarten V. de Hoop , Gabriel Peyré

Test-time training (TTT) methods explicitly update the weights of a model to adapt to the specific test instance, and they have found success in a variety of settings, including most recently language modeling and reasoning. To demystify…

Machine Learning · Computer Science 2026-02-24 Halil Alperen Gozeten , M. Emrullah Ildiz , Xuechen Zhang , Mahdi Soltanolkotabi , Marco Mondelli , Samet Oymak

We present models for embedding words in the context of surrounding words. Such models, which we refer to as token embeddings, represent the characteristics of a word that are specific to a given context, such as word sense, syntactic…

Computation and Language · Computer Science 2017-06-13 Lifu Tu , Kevin Gimpel , Karen Livescu

Causal transformer language models (LMs), such as GPT-3, typically require some form of positional encoding, such as positional embeddings. However, we show that LMs without any explicit positional encoding are still competitive with…

Computation and Language · Computer Science 2022-12-07 Adi Haviv , Ori Ram , Ofir Press , Peter Izsak , Omer Levy

Many neural network architectures are known to be Turing Complete, and can thus, in principle implement arbitrary algorithms. However, Transformers are unique in that they can implement gradient-based learning algorithms under simple…

Machine Learning · Computer Science 2024-06-05 Xiang Cheng , Yuxin Chen , Suvrit Sra

Task vectors, representing directions in model or activation spaces that encode task-specific behaviors, have emerged as a promising tool for steering large language models (LLMs). However, existing approaches typically require fine-tuning…

Computation and Language · Computer Science 2026-04-14 Feiyang Li , Yile Wang

This paper explores the elusive mechanism underpinning in-context learning in Large Language Models (LLMs). Our work provides a novel perspective by examining in-context learning via the lens of surface repetitions. We quantitatively…

Computation and Language · Computer Science 2024-02-22 Jianhao Yan , Jin Xu , Chiyu Song , Chenming Wu , Yafu Li , Yue Zhang

Task arithmetic has emerged as a simple yet powerful technique for model merging, enabling the combination of multiple finetuned models into one. Despite its empirical success, a clear theoretical explanation of why and when it works is…

Despite the empirical success of foundation models, we do not have a systematic characterization of the representations that these models learn. In this paper, we establish the contexture theory. It shows that a large class of…

Machine Learning · Computer Science 2025-05-06 Runtian Zhai , Kai Yang , Che-Ping Tsai , Burak Varici , Zico Kolter , Pradeep Ravikumar

In many real-world scenarios, data to train machine learning models becomes available over time. Unfortunately, these models struggle to continually learn new concepts without forgetting what has been learnt in the past. This phenomenon is…

Computation and Language · Computer Science 2023-01-16 Beyza Ermis , Giovanni Zappella , Martin Wistuba , Aditya Rawal , Cedric Archambeau

Transformers predict over a representation of a sequence. The same data can be written as bytes, characters, or subword tokens, and these representations may be lossless. Yet, under a fixed context window, they need not expose the same…

Machine Learning · Computer Science 2026-05-14 Amirmehdi Jafari Fesharaki , Mohammadamin Rami , Aslan Tchamkerten

Large language models (LLMs) exhibit a strong capacity for in-context learning: Given labeled examples, they can generate good predictions without parameter updates. However, many interactive settings go beyond static prediction to online…

Machine Learning · Computer Science 2026-05-12 Emile Anand , Abdullah Ateyeh , Xinyuan Cao , Max Dabagia