9.7k citations · 12.6k across the 24 of their papers we have counts for
24 papers
Linearity of Relation Decoding in Transformer Language Models
Evan Hernandez, Arnab Sen Sharma, Tal Haklay +5
Much of the knowledge encoded in transformer language models (LMs) may be expressed in terms of relations: relations between words and their synonyms, entities and their attributes…
Beyond Surface Statistics: Scene Representations in a Latent Diffusion Model
Yida Chen, Fernanda Viégas, Martin Wattenberg
Latent diffusion models (LDMs) exhibit an impressive ability to produce realistic images, yet the inner workings of these models remain mysterious. Even when trained purely on imag…
Inference-Time Intervention: Eliciting Truthful Answers from a Language Model
Kenneth Li, Oam Patel, Fernanda Viégas +2
We introduce Inference-Time Intervention (ITI), a technique designed to enhance the "truthfulness" of large language models (LLMs). ITI operates by shifting model activations durin…
The System Model and the User Model: Exploring AI Dashboard Design
Fernanda Viégas, Martin Wattenberg
This is a speculative essay on interface design and artificial intelligence. Recently there has been a surge of attention to chatbots based on large language models, including wide…
AttentionViz: A Global View of Transformer Attention
Catherine Yeh, Yida Chen, Aoyu Wu +3
Transformer models are revolutionizing machine learning, but their inner workings remain mysterious. In this work, we present a new visualization technique designed to help researc…
Investigating How Practitioners Use Human-AI Guidelines: A Case Study on the People + AI Guidebook
Nur Yildirim, Mahima Pushkarna, Nitesh Goyal +2
Artificial intelligence (AI) presents new challenges for the user experience (UX) of products and services. Recently, practitioner-facing resources and design guidelines have becom…
Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Task
Kenneth Li, Aspen K. Hopkins, David Bau +3
Language models show a surprising range of capabilities, but the source of their apparent competence is unclear. Do these networks just memorize a collection of surface statistics,…
Toy Models of Superposition
Nelson Elhage, Tristan Hume, Catherine Olsson +13
Neural networks often pack many unrelated concepts into a single neuron - a puzzling phenomenon known as 'polysemanticity' which makes interpretability much more challenging. This…
Interpreting a Machine Learning Model for Detecting Gravitational Waves
Mohammadtaher Safarzadeh, Asad Khan, E. A. Huerta +1
We describe a case study of translational research, applying interpretability techniques developed for computer vision to machine learning models used to search for and find gravit…
An Interpretability Illusion for BERT
Tolga Bolukbasi, Adam Pearce, Ann Yuan +4
We describe an "interpretability illusion" that arises when analyzing the BERT model. Activations of individual neurons in the network may spuriously appear to encode a single, sim…
"A cold, technical decision-maker": Can AI provide explainability, negotiability, and humanity?
Allison Woodruff, Yasmin Asare Anderson, Katherine Jameson Armstrong +8
Algorithmic systems are increasingly deployed to make decisions in many areas of people's lives. The shift from human to algorithmic decision-making has been accompanied by concern…
The What-If Tool: Interactive Probing of Machine Learning Models
James Wexler, Mahima Pushkarna, Tolga Bolukbasi +3
A key challenge in developing and deploying Machine Learning (ML) systems is understanding their performance across a wide range of inputs. To address this challenge, we created th…
Visualizing and Measuring the Geometry of BERT
Andy Coenen, Emily Reif, Ann Yuan +4
Transformer architectures show significant promise for natural language processing. Given that a single pretrained model can be fine-tuned to perform well on many different tasks,…
Neural Networks Trained on Natural Scenes Exhibit Gestalt Closure
Been Kim, Emily Reif, Martin Wattenberg +2
The Gestalt laws of perceptual organization, which describe how visual elements in an image are grouped and interpreted, have traditionally been thought of as innate despite their…
TensorFlow.js: Machine Learning for the Web and Beyond
Daniel Smilkov, Nikhil Thorat, Yannick Assogba +17
TensorFlow.js is a library for building and executing machine learning algorithms in JavaScript. TensorFlow.js models run in a web browser and in the Node.js environment. The libra…
Human-Centered Tools for Coping with Imperfect Algorithms during Medical Decision-Making
Carrie J. Cai, Emily Reif, Narayan Hegde +8
Machine learning (ML) is increasingly being used in image retrieval systems for medical decision making. One application of ML is to retrieve visually similar medical images from p…
GAN Lab: Understanding Complex Deep Generative Models using Interactive Visual Experimentation
Minsuk Kahng, Nikhil Thorat, Duen Horng Chau +2
Recent success in deep learning has generated immense interest among practitioners and students, inspiring many to learn about this new technology. While visual and interactive app…
Adversarial Spheres
Justin Gilmer, Luke Metz, Fartash Faghri +4
State of the art computer vision models have been shown to be vulnerable to small adversarial perturbations of the input. In other words, most images in the data distribution are b…
Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)
Been Kim, Martin Wattenberg, Justin Gilmer +4
The interpretation of deep learning models is a challenge due to their size, complexity, and often opaque internal state. In addition, many systems, such as image classifiers, oper…
Direct-Manipulation Visualization of Deep Networks
Daniel Smilkov, Shan Carter, D. Sculley +2
The recent successes of deep learning have led to a wave of interest from non-experts. Gaining an understanding of this technology, however, is difficult. While the theory is impor…
SmoothGrad: removing noise by adding noise
Daniel Smilkov, Nikhil Thorat, Been Kim +2
Explaining the output of a deep network remains a challenge. In the case of an image classifier, one type of explanation is to identify pixels that strongly influence the final dec…
Embedding Projector: Interactive Visualization and Interpretation of Embeddings
Daniel Smilkov, Nikhil Thorat, Charles Nicholson +3
Embeddings are ubiquitous in machine learning, appearing in recommender systems, NLP, and many other applications. Researchers and developers often need to explore the properties o…
Google's Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation
Melvin Johnson, Mike Schuster, Quoc V. Le +9
We propose a simple solution to use a single Neural Machine Translation (NMT) model to translate between multiple languages. Our solution requires no change in the model architectu…
TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
Martín Abadi, Ashish Agarwal, Paul Barham +37
TensorFlow is an interface for expressing machine learning algorithms, and an implementation for executing such algorithms. A computation expressed using TensorFlow can be executed…