English
Related papers

Related papers: The Impact of Depth on Compositional Generalizatio…

200 papers

In-context learning enables transformer models to generalize to new tasks based solely on input prompts, without any need for weight updates. However, existing training paradigms typically rely on large, unstructured datasets that are…

Large-scale pretraining instills large amounts of knowledge in deep neural networks. This, in turn, improves the generalization behavior of these models in downstream tasks. What exactly are the limits to the generalization benefits of…

Computation and Language · Computer Science 2022-12-23 A. Emin Orhan

For language models to generalize correctly to novel expressions, it is critical that they exploit access compositional meanings when this is justified. Even if we don't know what a "pelp" is, we can use our knowledge of numbers to…

Computation and Language · Computer Science 2025-09-25 Zhijin Guo , Chenhao Xue , Zhaozhen Xu , Hongbo Bo , Yuxuan Ye , Janet B. Pierrehumbert , Martha Lewis

Large-scale transformers achieve impressive results on program synthesis benchmarks, yet their true generalization capabilities remain obscured by data contamination and opaque training corpora. To rigorously assess whether models are truly…

Machine Learning · Computer Science 2026-05-01 Henrik Voigt , Michael Habeck , Joachim Giesen

Transformers have demonstrated impressive capabilities across various tasks, yet their performance on compositional problems remains a subject of debate. In this study, we investigate the internal mechanisms underlying Transformers'…

Computation and Language · Computer Science 2025-01-16 Zhongwang Zhang , Pengxiao Lin , Zhiwei Wang , Yaoyu Zhang , Zhi-Qin John Xu

Transformers have supplanted recurrent models in a large number of NLP tasks. However, the differences in their abilities to model different syntactic properties remain largely unknown. Past works suggest that LSTMs generalize very well on…

Computation and Language · Computer Science 2020-10-09 Satwik Bhattamishra , Kabir Ahuja , Navin Goyal

Despite their nearly universal adoption for large language models, the internal workings of transformers are not well understood. We aim to better understand the impact of removing or reorganizing information throughout the layers of a…

Computation and Language · Computer Science 2025-02-14 Qi Sun , Marc Pickett , Aakash Kumar Nain , Llion Jones

Recent diagnostic datasets on compositional generalization, such as SCAN (Lake and Baroni, 2018) and COGS (Kim and Linzen, 2020), expose severe problems in models trained from scratch on these datasets. However, in contrast to this poor…

Computation and Language · Computer Science 2023-11-09 Xiang Zhou , Yichen Jiang , Mohit Bansal

Deep transformer models have pushed performance on NLP tasks to new limits, suggesting sophisticated treatment of complex linguistic inputs, such as phrases. However, we have limited understanding of how these models handle representation…

Computation and Language · Computer Science 2020-10-15 Lang Yu , Allyson Ettinger

While mainstream machine learning methods are known to have limited ability to compositionally generalize, new architectures and techniques continue to be proposed to address this limitation. We investigate state-of-the-art techniques and…

Computation and Language · Computer Science 2021-09-23 Daniel Furrer , Marc van Zee , Nathan Scales , Nathanael Schärli

Transformer language models have received widespread public attention, yet their generated text is often surprising even to NLP researchers. In this survey, we discuss over 250 recent studies of English language model behavior before…

Computation and Language · Computer Science 2023-08-29 Tyler A. Chang , Benjamin K. Bergen

Pretraining on large, semantically rich datasets is key for developing language models. Surprisingly, recent studies have shown that even synthetic data, generated procedurally through simple semantic-free algorithms, can yield some of the…

Machine Learning · Computer Science 2025-05-29 Zachary Shinnick , Liangze Jiang , Hemanth Saratchandran , Anton van den Hengel , Damien Teney

Transformers trained on huge text corpora exhibit a remarkable set of capabilities, e.g., performing basic arithmetic. Given the inherent compositional nature of language, one can expect the model to learn to compose these capabilities,…

Machine Learning · Computer Science 2024-02-07 Rahul Ramesh , Ekdeep Singh Lubana , Mikail Khona , Robert P. Dick , Hidenori Tanaka

Transformer-based models excel in various tasks but their generalization capabilities, especially in arithmetic reasoning, remain incompletely understood. Arithmetic tasks provide a controlled framework to explore these capabilities, yet…

Machine Learning · Computer Science 2025-08-07 Xingcheng Xu , Zibo Zhao , Haipeng Zhang , Yanqing Yang

Transformer networks have seen great success in natural language processing and machine vision, where task objectives such as next word prediction and image classification benefit from nuanced context sensitivity across high-dimensional…

Machine Learning · Computer Science 2022-12-13 Yuxuan Li , James L. McClelland

Compositional understanding is crucial for human intelligence, yet it remains unclear whether contemporary vision models exhibit it. The dominant machine learning paradigm is built on the premise that scaling data and model sizes will…

Machine Learning · Computer Science 2025-07-10 Arnas Uselis , Andrea Dittadi , Seong Joon Oh

Model-based methods and deep neural networks have both been tremendously successful paradigms in machine learning. In model-based methods, problem domain knowledge can be built into the constraints of the model, typically at the expense of…

Machine Learning · Computer Science 2014-11-21 John R. Hershey , Jonathan Le Roux , Felix Weninger

Large Language Models (LLMs) often exhibit a gap between their internal knowledge and their explicit linguistic outputs. In this report, we empirically investigate whether Looped Transformers (LTs)--architectures that increase computational…

Computation and Language · Computer Science 2026-01-16 Guanxu Chen , Dongrui Liu , Jing Shao

Transformers trained on natural language data have been shown to learn its hierarchical structure and generalize to sentences with unseen syntactic structures without explicitly encoding any structural bias. In this work, we investigate…

Computation and Language · Computer Science 2025-03-18 Kabir Ahuja , Vidhisha Balachandran , Madhur Panwar , Tianxing He , Noah A. Smith , Navin Goyal , Yulia Tsvetkov

Understanding what defines a good representation in large language models (LLMs) is fundamental to both theoretical understanding and practical applications. In this paper, we investigate the quality of intermediate representations in…

Machine Learning · Computer Science 2024-12-13 Oscar Skean , Md Rifat Arefin , Yann LeCun , Ravid Shwartz-Ziv