English
Related papers

Related papers: Pause Tokens Strictly Increase the Expressivity of…

200 papers

Chain-of-thought and more broadly test-time compute are known to augment the expressive capabilities of language models and have led to major innovations in reasoning. Motivated by this success, this paper explores latent chain-of-thought…

Machine Learning · Computer Science 2026-05-20 Carson Dudley , Samet Oymak

Transformers update token representations through multi-head attention and residual connections as $X \leftarrow X + \sum_{i} P^{(i)}XW_{V_i}W_{o_i}$, where $P^{(i)}$ is the softmax attention matrix in head $i$. We propose replacing a…

Machine Learning · Computer Science 2026-05-18 Yuchong Zhang , Vardan Papyan

Reinforcement-learning-based post-training has become a key approach for improving the reasoning ability of large language models, but its token-level learning signals remain poorly understood. This work studies their heterogeneity through…

Computation and Language · Computer Science 2026-05-11 Gengyang Li , Zheng-Fan Wu , Siqi Bao , Yunfang Wu

Transformer is a powerful model for text understanding. However, it is inefficient due to its quadratic complexity to input sequence length. Although there are many methods on Transformer acceleration, they are still either inefficient on…

Computation and Language · Computer Science 2021-09-07 Chuhan Wu , Fangzhao Wu , Tao Qi , Yongfeng Huang , Xing Xie

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

We conduct a systematic study of the approximation properties of Transformer for sequence modeling with long, sparse and complicated memory. We investigate the mechanisms through which different components of Transformer, such as the…

Machine Learning · Computer Science 2024-10-31 Mingze Wang , Weinan E

There is a growing interest in the ability of neural networks to execute algorithmic tasks (e.g., arithmetic, summary statistics, and sorting). The goal of this work is to better understand the role of attention in Transformers for…

Machine Learning · Computer Science 2025-06-11 Artur Back de Luca , George Giapitzakis , Shenghao Yang , Petar Veličković , Kimon Fountoulakis

Symbolic music generation aims to generate music scores automatically. A recent trend is to use Transformer or its variants in music generation, which is, however, suboptimal, because the full attention cannot efficiently model the…

Sound · Computer Science 2022-11-01 Botao Yu , Peiling Lu , Rui Wang , Wei Hu , Xu Tan , Wei Ye , Shikun Zhang , Tao Qin , Tie-Yan Liu

Vision transformers have established a precedent of patchifying images into uniformly-sized chunks before processing. We hypothesize that this design choice may limit models in learning comprehensive and compositional representations from…

Computer Vision and Pattern Recognition · Computer Science 2025-05-20 Neha Kalibhat , Priyatham Kattakinda , Sumit Nawathe , Arman Zarei , Nikita Seleznev , Samuel Sharpe , Senthil Kumar , Soheil Feizi

Chain-of-thought (CoT) rationale enables language models to use additional task-related text for problem-solving, benefiting not only from detailed reasoning steps but also from the expanded computational space of longer inputs. Prior work…

Computation and Language · Computer Science 2025-09-30 Yoonna Jang , Kisu Yang , Isabelle Augenstein

Looped Transformers provide advantages in parameter efficiency, computational capabilities, and generalization for reasoning tasks. However, their expressive power regarding function approximation remains underexplored. In this paper, we…

Machine Learning · Computer Science 2025-06-06 Kevin Xu , Issei Sato

Chain-of-thought (CoT) reasoning improves large language models (LLMs) on difficult tasks, but it also makes inference expensive because every intermediate step must be generated as a discrete token. Latent reasoning reduces visible token…

Computation and Language · Computer Science 2026-05-11 Xuan Li , Yining Wang , Yuchen Liu , Guanjun Liu , Delai Qiu , Shengping Liu , Jiaen Liang , Wei Huang , Jun Yu , Junnan Zhu

Transformers have exhibited exceptional capabilities in sequence modeling tasks, leveraging self-attention and in-context learning. Critical to this success are induction heads, attention circuits that enable copying tokens based on their…

Machine Learning · Computer Science 2025-09-11 Francesco D'Angelo , Francesco Croce , Nicolas Flammarion

We introduce token-consistent stochastic layers in vision transformers, without causing any severe drop in performance. The added stochasticity improves network calibration, robustness and strengthens privacy. We use linear layers with…

Computer Vision and Pattern Recognition · Computer Science 2022-07-15 Nikola Popovic , Danda Pani Paudel , Thomas Probst , Luc Van Gool

Metaphor requires a language model to resolve a token whose contextual meaning diverges from its basic literal sense. Understanding how transformer models organize this reinterpretation across depth remains an open problem in mechanistic…

Computation and Language · Computer Science 2026-05-21 Lawhori Chakrabarti , Jennifer Johnson-Leung , Bert Baumgaertner , Aleksandar Vakanski , Min Xian , Boyu Zhang

Transformers have demonstrated great success in numerous domains including natural language processing and bioinformatics. This success stems from the use of the attention mechanism by these models in order to represent and propagate…

Machine Learning · Computer Science 2025-02-10 Nathaniel Tomczak , Sanmukh Kuppannagari

Since its introduction in 2017, Transformer has emerged as the leading neural network architecture, catalyzing revolutionary advancements in many AI disciplines. The key innovation in Transformer is a Self-Attention (SA) mechanism designed…

Machine Learning · Computer Science 2024-02-05 Yueyao Yu , Yin Zhang

Recent theoretical work has identified surprisingly simple reasoning problems, such as checking if two nodes in a graph are connected or simulating finite-state machines, that are provably unsolvable by standard transformers that answer…

Machine Learning · Computer Science 2024-04-15 William Merrill , Ashish Sabharwal

Current neural architectures lack a principled way to handle interchangeable tokens, i.e., symbols that are semantically equivalent yet distinguishable, such as bound variables. As a result, models trained on fixed vocabularies often…

Machine Learning · Computer Science 2026-02-02 İlker Işık , Wenchao Li

It has been observed that transformers with greater depth (that is, more layers) have more capabilities, but can we establish formally which capabilities are gained? We answer this question with a theoretical proof followed by an empirical…

Computation and Language · Computer Science 2026-01-21 Andy Yang , Michaël Cadilhac , David Chiang