English
Related papers

Related papers: Implicit Geometry of Next-token Prediction: From L…

200 papers

Autoregressive language models, pretrained using large text corpora to do well on next word prediction, have been successful at solving many downstream tasks, even with zero-shot usage. However, there is little theoretical understanding of…

Computation and Language · Computer Science 2021-04-15 Nikunj Saunshi , Sadhika Malladi , Sanjeev Arora

Interpretability benefits the theoretical understanding of representations. Existing word embeddings are generally dense representations. Hence, the meaning of latent dimensions is difficult to interpret. This makes word embeddings like a…

Computation and Language · Computer Science 2023-06-27 Minxue Xia , Hao Zhu

Topic models have evolved from conventional Bayesian probabilistic models to recent Neural Topic Models (NTMs). Although NTMs have shown promising performance when trained and tested on a specific corpus, their generalisation ability across…

Computation and Language · Computer Science 2024-06-14 Xiaohao Yang , He Zhao , Dinh Phung , Lan Du

Neural Collapse is a phenomenon that helps identify sparse and low rank structures in deep classifiers. Recent work has extended the definition of neural collapse to regression problems, albeit only measuring the phenomenon at the last…

Machine Learning · Computer Science 2026-03-26 Akshay Rangamani , Altay Unal

Translation into morphologically-rich languages challenges neural machine translation (NMT) models with extremely sparse vocabularies where atomic treatment of surface forms is unrealistic. This problem is typically addressed by either…

Computation and Language · Computer Science 2020-02-28 Duygu Ataman , Wilker Aziz , Alexandra Birch

Recurrent neural networks have been very successful at predicting sequences of words in tasks such as language modeling. However, all such models are based on the conventional classification framework, where the model is trained against…

Machine Learning · Computer Science 2017-03-14 Hakan Inan , Khashayar Khosravi , Richard Socher

Word embeddings are a powerful natural language processing technique, but they are extremely difficult to interpret. To enable interpretable NLP models, we create vectors where each dimension is inherently interpretable. By inherently…

Computation and Language · Computer Science 2021-09-29 Adly Templeton

We systematically investigate multi-token prediction (MTP) capabilities within LLMs pre-trained for next-token prediction (NTP). We first show that such models inherently possess MTP capabilities via numerical marginalization over…

Computation and Language · Computer Science 2025-02-14 Somesh Mehra , Javier Alonso Garcia , Lukas Mauch

Recent works have argued that high-level semantic concepts are encoded "linearly" in the representation space of large language models. In this work, we study the origins of such linear representations. To that end, we introduce a simple…

Computation and Language · Computer Science 2024-03-07 Yibo Jiang , Goutham Rajendran , Pradeep Ravikumar , Bryon Aragam , Victor Veitch

Large Language Models (LLMs), such as GPT, are considered to learn the latent distributions within large-scale web-crawl datasets and accomplish natural language processing (NLP) tasks by predicting the next token. However, this mechanism…

Computation and Language · Computer Science 2025-02-04 Kun-Peng Ning , Jia-Yu Yao , Yu-Yang Liu , Mu-Nan Ning , Li Yuan

Neural models learn representations of high-dimensional data on low-dimensional manifolds. Multiple factors, including stochasticities in the training process, model architectures, and additional inductive biases, may induce different…

Machine Learning · Computer Science 2025-12-02 Hanlin Yu , Berfin Inal , Georgios Arvanitidis , Soren Hauberg , Francesco Locatello , Marco Fumero

Multilingual language models (LMs) promise broader NLP access, yet current systems deliver uneven performance across the world's languages. This survey examines why these gaps persist and whether they reflect intrinsic linguistic difficulty…

Computation and Language · Computer Science 2026-04-13 Chen Shani , Yuval Reif , Nathan Roll , Dan Jurafsky , Ekaterina Shutova

With the recent success of pre-trained models in NLP, a significant focus was put on interpreting their representations. One of the most prominent approaches is structural probing (Hewitt and Manning, 2019), where a linear projection of…

Computation and Language · Computer Science 2021-06-25 Tomasz Limisiewicz , David Mareček

Pretraining Neural Language Models (NLMs) over a large corpus involves chunking the text into training examples, which are contiguous text segments of sizes processable by the neural architecture. We highlight a bias introduced by this…

Computation and Language · Computer Science 2022-03-22 Yoav Levine , Noam Wies , Daniel Jannai , Dan Navon , Yedid Hoshen , Amnon Shashua

It has been long known that sparsity is an effective inductive bias for learning efficient representation of data in vectors with fixed dimensionality, and it has been explored in many areas of representation learning. Of particular…

Computation and Language · Computer Science 2021-04-20 Victor Prokhorov , Yingzhen Li , Ehsan Shareghi , Nigel Collier

The static ``train then deploy" paradigm fundamentally limits Large Language Models (LLMs) from dynamically adapting their weights in response to continuous streams of new information inherent in real-world tasks. Test-Time Training (TTT)…

Machine Learning · Computer Science 2026-04-08 Guhao Feng , Shengjie Luo , Kai Hua , Ge Zhang , Di He , Wenhao Huang , Tianle Cai

Modern NLP models rely heavily on engineered features, which often combine word and contextual information into complex lexical features. Such combination results in large numbers of features, which can lead to over-fitting. We present a…

Computation and Language · Computer Science 2016-04-05 Mo Yu , Mark Dredze , Raman Arora , Matthew Gormley

While the success of pre-trained language models has largely eliminated the need for high-quality static word vectors in many NLP applications, such vectors continue to play an important role in tasks where words need to be modelled in the…

Computation and Language · Computer Science 2021-05-18 Na Li , Zied Bouraoui , Jose Camacho Collados , Luis Espinosa-Anke , Qing Gu , Steven Schockaert

Large neural networks are typically trained for a fixed computational budget, creating a rigid trade-off between performance and efficiency that is ill-suited for deployment in resource-constrained or dynamic environments. Existing…

Machine Learning · Computer Science 2026-03-05 Paulius Rauba , Mihaela van der Schaar

Phonotactic constraints can be employed to distinguish languages by representing a speech utterance as a multinomial distribution or phone events. In the present study, we propose a new learning mechanism based on subspace-based…

Sound · Computer Science 2022-03-30 Hung-Shin Lee , Yu Tsao , Shyh-Kang Jeng , Hsin-Min Wang