English
Related papers

Related papers: Obtaining Faithful Interpretations from Compositio…

200 papers

This paper aims to understand how neural networks learn algorithmic reasoning by addressing two questions: How faithful are learned algorithms when they are effective, and why do neural networks fail to learn effective algorithms otherwise?…

Artificial Intelligence · Computer Science 2025-12-09 Lucas Saldyt , Subbarao Kambhampati

Neural network models have achieved state-of-the-art performances in a wide range of natural language processing (NLP) tasks. However, a long-standing criticism against neural network models is the lack of interpretability, which not only…

Computation and Language · Computer Science 2021-10-26 Xiaofei Sun , Diyi Yang , Xiaoya Li , Tianwei Zhang , Yuxian Meng , Han Qiu , Guoyin Wang , Eduard Hovy , Jiwei Li

We propose Modal Logical Neural Networks (MLNNs), a neurosymbolic framework that integrates deep learning with the formal semantics of modal logic, enabling reasoning about necessity and possibility. Drawing on Kripke semantics, we…

Machine Learning · Computer Science 2026-02-13 Antonin Sulc

A central goal for mechanistic interpretability has been to identify the right units of analysis in large language models (LLMs) that causally explain their outputs. While early work focused on individual neurons, evidence that neurons…

Computation and Language · Computer Science 2026-05-05 Or Shafran , Atticus Geiger , Mor Geva

Deep neural networks (DNNs) are powerful black-box predictors that have achieved impressive performance on a wide variety of tasks. However, their accuracy comes at the cost of intelligibility: it is usually unclear how they make their…

Machine Learning · Computer Science 2021-10-26 Rishabh Agarwal , Levi Melnick , Nicholas Frosst , Xuezhou Zhang , Ben Lengerich , Rich Caruana , Geoffrey Hinton

Transformer large language models (LLMs) have sparked admiration for their exceptional performance on tasks that demand intricate multi-step reasoning. Yet, these models simultaneously show failures on surprisingly trivial problems. This…

Compositionality is a pivotal property of symbolic reasoning. However, how well recent neural models capture compositionality remains underexplored in the symbolic reasoning tasks. This study empirically addresses this question by…

Computation and Language · Computer Science 2023-02-16 Keito Kudo , Yoichi Aoki , Tatsuki Kuribayashi , Ana Brassard , Masashi Yoshikawa , Keisuke Sakaguchi , Kentaro Inui

The compositional generalization abilities of neural models have been sought after for human-like linguistic competence. The popular method to evaluate such abilities is to assess the models' input-output behavior. However, that does not…

Computation and Language · Computer Science 2025-02-24 Ryoma Kumon , Hitomi Yanaka

The mathematical representation of semantics is a key issue for Natural Language Processing (NLP). A lot of research has been devoted to finding ways of representing the semantics of individual words in vector spaces. Distributional…

Computation and Language · Computer Science 2014-11-13 Karl Moritz Hermann

Deep neural networks have proven remarkably effective at solving many classification problems, but have been criticized recently for two major weaknesses: the reasons behind their predictions are uninterpretable, and the predictions…

Machine Learning · Computer Science 2017-11-28 Andrew Slavin Ross , Finale Doshi-Velez

Visual Question Answering (VQA) models have achieved significant success in recent times. Despite the success of VQA models, they are mostly black-box models providing no reasoning about the predicted answer, thus raising questions for…

Computer Vision and Pattern Recognition · Computer Science 2021-05-18 Nihar Bendre , Kevin Desai , Peyman Najafirad

Recurrent Neural Networks (RNNs) have achieved remarkable performance on a range of tasks. A key step to further empowering RNN-based approaches is improving their explainability and interpretability. In this work we present MEME: a model…

Machine Learning · Computer Science 2021-04-15 Dmitry Kazhdan , Botty Dimanov , Mateja Jamnik , Pietro Liò

Compositional and relational learning is a hallmark of human intelligence, but one which presents challenges for neural models. One difficulty in the development of such models is the lack of benchmarks with clear compositional and…

Machine Learning · Computer Science 2020-07-09 Tim Klinger , Dhaval Adjodah , Vincent Marois , Josh Joseph , Matthew Riemer , Alex 'Sandy' Pentland , Murray Campbell

This paper addresses the task of designing a modular neural network architecture that jointly solves different tasks. As an example we use the tasks of depth estimation and semantic segmentation given a single RGB image. The main focus of…

Computer Vision and Pattern Recognition · Computer Science 2018-09-27 Omid Hosseini Jafari , Oliver Groth , Alexander Kirillov , Michael Ying Yang , Carsten Rother

Computational modeling plays an increasingly important role in neuroscience, highlighting the philosophical question of how computational models explain. In the context of neural network models for neuroscience, concerns have been raised…

Neurons and Cognition · Quantitative Biology 2021-04-15 Rosa Cao , Daniel Yamins

We ask whether the neural network interpretation methods can be fooled via adversarial model manipulation, which is defined as a model fine-tuning step that aims to radically alter the explanations without hurting the accuracy of the…

Machine Learning · Computer Science 2019-11-04 Juyeon Heo , Sunghwan Joo , Taesup Moon

State of the art Deep Neural Networks (DNN) can now achieve above human level accuracy on image classification tasks. However their outstanding performances come along with a complex inference mechanism making them arduously interpretable…

Machine Learning · Computer Science 2019-11-07 Fei Wu , Thomas Michel , Alexandre Briot

Compositional Explanations is a method for identifying logical formulas of concepts that approximate the neurons' behavior. However, these explanations are linked to the small spectrum of neuron activations (i.e., the highest ones) used to…

Machine Learning · Computer Science 2023-10-31 Biagio La Rosa , Leilani H. Gilpin , Roberto Capobianco

Neural networks leverage robust internal representations in order to generalise. Learning them is difficult, and often requires a large training set that covers the data distribution densely. We study a common setting where our task is not…

We propose a new interpretability method for neural networks, which is based on a novel mathematico-philosophical theory of reasons. Our method computes a vector for each neuron, called its reasons vector. We then can compute how strongly…

Machine Learning · Computer Science 2025-05-21 Levin Hornischer , Hannes Leitgeb