English
Related papers

Related papers: Obtaining Faithful Interpretations from Compositio…

200 papers

Language models' (LMs) proficiency in handling deterministic symbolic reasoning and rule-based tasks remains limited due to their dependency implicit learning on textual data. To endow LMs with genuine rule comprehension abilities, we…

Computation and Language · Computer Science 2024-03-12 Yixuan Weng , Minjun Zhu , Fei Xia , Bin Li , Shizhu He , Kang Liu , Jun Zhao

Mechanistic interpretability aims to explain neural model behaviour by reverse-engineering learned computational structure into human-understandable components. Without a formal framework, however, mechanistic explanations cannot be…

Machine Learning · Computer Science 2026-05-12 Ward Gauderis , Thomas Dooms , Steven T. Holmer , Kola Ayonrinde , Geraint A. Wiggins

We investigate two specific manifestations of compositionality in Neural Machine Translation (NMT) : (1) Productivity - the ability of the model to extend its predictions beyond the observed length in training data and (2) Systematicity -…

Computation and Language · Computer Science 2019-12-17 Vikas Raunak , Vaibhav Kumar , Florian Metze

The increasing use of complex machine learning models in education has led to concerns about their interpretability, which in turn has spurred interest in developing explainability techniques that are both faithful to the model's inner…

Machine Learning · Computer Science 2025-05-13 Juan D. Pinto , Luc Paquette

Compositional, structured models are appealing because they explicitly decompose problems and provide interpretable intermediate outputs that give confidence that the model is not simply latching onto data artifacts. Learning these models…

Computation and Language · Computer Science 2021-04-06 Nitish Gupta , Sameer Singh , Matt Gardner , Dan Roth

Neuron Interpretation has gained traction in the field of interpretability, and have provided fine-grained insights into what a model learns and how language knowledge is distributed amongst its different components. However, the lack of…

Computation and Language · Computer Science 2023-11-07 Yimin Fan , Fahim Dalvi , Nadir Durrani , Hassan Sajjad

Visual question answering requires high-order reasoning about an image, which is a fundamental capability needed by machine systems to follow complex directives. Recently, modular networks have been shown to be an effective framework for…

Computer Vision and Pattern Recognition · Computer Science 2019-01-24 David Mascharka , Philip Tran , Ryan Soklaski , Arjun Majumdar

Predicting multiple real-world tasks in a single model often requires a particularly diverse feature space. Multimodal (MM) models aim to extract the synergistic predictive potential of multiple data types to create a shared feature space…

Machine Learning · Computer Science 2023-11-07 Vinitra Swamy , Malika Satayeva , Jibril Frej , Thierry Bossy , Thijs Vogels , Martin Jaggi , Tanja Käser , Mary-Anne Hartley

The collaborative reasoning for understanding each image-question pair is very critical but under-explored for an interpretable Visual Question Answering (VQA) system. Although very recent works also tried the explicit compositional…

Computer Vision and Pattern Recognition · Computer Science 2018-04-03 Qingxing Cao , Xiaodan Liang , Bailing Li , Guanbin Li , Liang Lin

Neural networks have succeeded in many reasoning tasks. Empirically, these tasks require specialized network structures, e.g., Graph Neural Networks (GNNs) perform well on many such tasks, but less structured networks fail. Theoretically,…

Machine Learning · Computer Science 2020-02-18 Keyulu Xu , Jingling Li , Mozhi Zhang , Simon S. Du , Ken-ichi Kawarabayashi , Stefanie Jegelka

In traditional software programs, it is easy to trace program logic from variables back to input, apply assertion statements to block erroneous behavior, and compose programs together. Although deep learning programs have demonstrated…

Machine Learning · Computer Science 2021-10-27 Mike Wu , Noah Goodman , Stefano Ermon

The current understanding of deep neural networks can only partially explain how input structure, network parameters and optimization algorithms jointly contribute to achieve the strong generalization power that is typically observed in…

Machine Learning · Computer Science 2021-01-28 Francesco Craighero , Fabrizio Angaroni , Alex Graudenzi , Fabio Stella , Marco Antoniotti

Large language models (LLMs) have shown an impressive ability to perform tasks believed to require thought processes. When the model does not document an explicit thought process, it becomes difficult to understand the processes occurring…

Computation and Language · Computer Science 2024-06-21 Yuval Shalev , Amir Feder , Ariel Goldstein

Neurons are the fundamental building blocks of deep neural networks, and their interconnections allow AI to achieve unprecedented results. Motivated by the goal of understanding how neurons encode information, compositional explanations…

Computer Vision and Pattern Recognition · Computer Science 2025-11-27 Biagio La Rosa , Leilani H. Gilpin

Numerical reasoning skills are essential for complex question answering (CQA) over text. It requires opertaions including counting, comparison, addition and subtraction. A successful approach to CQA on text, Neural Module Networks (NMNs),…

Computation and Language · Computer Science 2021-09-07 Xiao-Yu Guo , Yuan-Fang Li , Gholamreza Haffari

Deep Neural Networks (DNNs) deliver impressive performance but their black-box nature limits deployment in high-stakes domains requiring transparency. We introduce Compositional Function Networks (CFNs), a novel framework that builds…

Machine Learning · Computer Science 2025-08-01 Fang Li

A formal description of the compositionality of neural networks is associated directly with the formal grammar-structure of the objects it seeks to represent. This formal grammar-structure specifies the kind of components that make up an…

Machine Learning · Computer Science 2020-10-07 Sai Raam Venkatraman , Ankit Anand , S. Balasubramanian , R. Raghunatha Sarma

Humans are able to explain their reasoning. On the contrary, deep neural networks are not. This paper attempts to bridge this gap by introducing a new way to design interpretable neural networks for classification, inspired by physiological…

Machine Learning · Statistics 2017-11-20 Shane Barratt

The extent to which neural networks are able to acquire and represent symbolic rules remains a key topic of research and debate. Much current work focuses on the impressive capabilities of large language models, as well as their often…

Machine Learning · Computer Science 2025-06-11 Anna Langedijk , Jaap Jumelet , Willem Zuidema

Complex reasoning over text requires understanding and chaining together free-form predicates and logical connectives. Prior work has largely tried to do this either symbolically or with black-box transformers. We present a middle ground…

Computation and Language · Computer Science 2021-06-08 Jiangming Liu , Matt Gardner , Shay B. Cohen , Mirella Lapata