English
Related papers

Related papers: Where is the signal in tokenization space?

200 papers

While Large Language Models (LLMs) have shown exceptional performance in various tasks, one of their most prominent drawbacks is generating inaccurate or false information with a confident tone. In this paper, we provide evidence that the…

Computation and Language · Computer Science 2023-10-18 Amos Azaria , Tom Mitchell

Unsupervised text representation learning (TRL) is a fundamental task in natural language processing, which is beneficial for improving search and recommendations with the web's unlabeled texts. A recent empirical study finds that the…

Computation and Language · Computer Science 2025-10-14 Ruize An , Richong Zhang , Zhijie Nie , Zhanyu Wu , Yanzhao Zhang , Dingkun Long

Training large language models (LLMs) from scratch can yield models with unique functionalities and strengths, but it is costly and often leads to redundant capabilities. A more cost-effective alternative is to fuse existing pre-trained…

Computation and Language · Computer Science 2025-09-23 Runjia Zeng , James Chenhao Liang , Cheng Han , Zhiwen Cao , Jiahao Liu , Xiaojun Quan , Yingjie Victor Chen , Lifu Huang , Tong Geng , Qifan Wang , Dongfang Liu

As language models scale, their performance improves dramatically across a wide range of tasks, but so does their tendency to memorize and regurgitate parts of their training data verbatim. This tradeoff poses serious legal, ethical, and…

Machine Learning · Computer Science 2025-05-28 Parjanya Prajakta Prashant , Kaustubh Ponkshe , Babak Salimi

Text representation plays a critical role in tasks like clustering, retrieval, and other downstream applications. With the emergence of large language models (LLMs), there is increasing interest in harnessing their capabilities for this…

Computation and Language · Computer Science 2025-12-25 Yeqin Zhang , Yizheng Zhao , Chen Hu , Binxing Jiao , Daxin Jiang , Ruihang Miao , Cam-Tu Nguyen

Large Language Models (LLMs) are a powerful tool for statistical text analysis, with derived sequences of next-token probability distributions offering a wealth of information. Extracting this signal typically relies on metrics such as…

Large Language Models (LLMs) are adept at text manipulation -- tasks such as machine translation and text summarization. However, these models can also be prone to hallucination, which can be detrimental to the faithfulness of any answers…

Computation and Language · Computer Science 2024-04-04 Priyesh Vakharia , Devavrat Joshi , Meenal Chavan , Dhananjay Sonawane , Bhrigu Garg , Parsa Mazaheri

The automated detection of hallucinations and training data contamination is pivotal to the safe deployment of Large Language Models (LLMs). These tasks are particularly challenging in settings where no access to model internals is…

Machine Learning · Computer Science 2025-10-01 Guy Bar-Shalom , Fabrizio Frasca , Derek Lim , Yoav Gelberg , Yftah Ziser , Ran El-Yaniv , Gal Chechik , Haggai Maron

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

Representing token embeddings as probability distributions over learned manifolds allows for more flexible contextual inference, reducing representational rigidity while enhancing semantic granularity. Comparative evaluations demonstrate…

Computation and Language · Computer Science 2025-04-25 Christopher Nightingale , Dominic Lavington , Jonathan Thistlethwaite , Sebastian Penhaligon , Thomas Belinski , David Boldo

Transforming a large language model (LLM) into a Vision-Language Model (VLM) can be achieved by mapping the visual tokens from a vision encoder into the embedding space of an LLM. Intriguingly, this mapping can be as simple as a shallow MLP…

Computer Vision and Pattern Recognition · Computer Science 2026-02-26 Benno Krojer , Shravan Nayak , Oscar Mañas , Vaibhav Adlakha , Desmond Elliott , Siva Reddy , Marius Mosbach

We conduct a large scale empirical investigation of contextualized number prediction in running text. Specifically, we consider two tasks: (1)masked number prediction-predicting a missing numerical value within a sentence, and (2)numerical…

Computation and Language · Computer Science 2020-11-17 Daniel Spokoyny , Taylor Berg-Kirkpatrick

Despite it being the cornerstone of BPE, the most common tokenization algorithm, the importance of compression in the tokenization process is still unclear. In this paper, we argue for the theoretical importance of compression, that can be…

Computation and Language · Computer Science 2024-06-25 Omer Goldman , Avi Caciularu , Matan Eyal , Kris Cao , Idan Szpektor , Reut Tsarfaty

Autoregressive large language models (LLMs) generate text token-by-token, requiring n forward passes to produce a sequence of length n. Recent work, Exploring the Latent Capacity of LLMs for One-Step Text Reconstruction (Mezentsev and…

Machine Learning · Computer Science 2026-02-23 Ivan Bondarenko , Egor Palkin , Fedor Tikunov

Large Language Models (LLMs) have demonstrated impressive performance on multimodal tasks, without any multimodal finetuning. They are the building block for Large Multimodal Models, yet, we still lack a proper understanding of their…

Computer Vision and Pattern Recognition · Computer Science 2024-10-08 Mustafa Shukor , Matthieu Cord

The scarcity of large parallel corpora is an important obstacle for neural machine translation. A common solution is to exploit the knowledge of language models (LM) trained on abundant monolingual data. In this work, we propose a novel…

Computation and Language · Computer Science 2020-10-27 Christos Baziotis , Barry Haddow , Alexandra Birch

Large language models (LLMs) have exhibited impressive competence in various tasks, but their internal mechanisms on mathematical problems are still under-explored. In this paper, we study a fundamental question: how language models encode…

Computation and Language · Computer Science 2024-11-15 Fangwei Zhu , Damai Dai , Zhifang Sui

In this study, the output of large language models (LLM) is considered an information source generating an unlimited sequence of symbols drawn from a finite alphabet. Given the probabilistic nature of modern LLMs, we assume a probabilistic…

Computation and Language · Computer Science 2026-02-24 Marco Scharringhausen

Computing next-token likelihood ratios between two language models (LMs) is a standard task in training paradigms such as knowledge distillation. Since this requires both models to share the same probability space, it becomes challenging…

Computation and Language · Computer Science 2026-05-07 Buu Phan , Ashish Khisti , Karen Ullrich

LLMs process text as sequences of tokens that roughly correspond to words, where less common words are represented by multiple tokens. However, individual tokens are often semantically unrelated to the meanings of the words/concepts they…

Computation and Language · Computer Science 2024-10-14 Sheridan Feucht , David Atkinson , Byron Wallace , David Bau