18 citations · 22 across the 9 of their papers we have counts for
10 papers
Line of Sight: On Linear Representations in VLLMs
Achyuta Rajaram, Sarah Schwettmann, Jacob Andreas +1
Language models can be equipped with multimodal capabilities by fine-tuning on embeddings of visual inputs. But how do such multimodal models represent images in their hidden activ…
Nearest Neighbor Normalization Improves Multimodal Retrieval
Neil Chowdhury, Franklin Wang, Sumedh Shenoy +3
Multimodal models leverage large-scale pre-training to achieve strong but still imperfect performance on tasks such as image captioning, visual question answering, and cross-modal…
Automatic Discovery of Visual Circuits
Achyuta Rajaram, Neil Chowdhury, Antonio Torralba +2
To date, most discoveries of network subcomponents that implement human-interpretable computations in deep vision models have involved close study of single units and large amounts…
A Multimodal Automated Interpretability Agent
Tamar Rott Shaham, Sarah Schwettmann, Franklin Wang +4
This paper describes MAIA, a Multimodal Automated Interpretability Agent. MAIA is a system that uses neural models to automate neural model understanding tasks like feature interpr…
An Alternative to Regulation: The Case for Public AI
Nicholas Vincent, David Bau, Sarah Schwettmann +1
Can governments build AI? In this paper, we describe an ongoing effort to develop ``public AI'' -- publicly accessible AI models funded, provisioned, and governed by governments or…
Multimodal Neurons in Pretrained Text-Only Transformers
Sarah Schwettmann, Neil Chowdhury, Samuel Klein +2
Language models demonstrate remarkable capacity to generalize representations learned in one modality to downstream tasks in other modalities. Can we trace this ability to individu…
FIND: A Function Description Benchmark for Evaluating Interpretability Methods
Sarah Schwettmann, Tamar Rott Shaham, Joanna Materzynska +5
Labeling neural network submodules with human-legible descriptions is useful for many downstream tasks: such descriptions can surface failures, guide interventions, and perhaps eve…
Natural Language Descriptions of Deep Visual Features
Evan Hernandez, Sarah Schwettmann, David Bau +3
Some neurons in deep networks specialize in recognizing highly specific perceptual, structural, or semantic features of inputs. In computer vision, techniques exist for identifying…
Toward a Visual Concept Vocabulary for GAN Latent Space
Sarah Schwettmann, Evan Hernandez, David Bau +3
A large body of recent work has identified transformations in the latent spaces of generative adversarial networks (GANs) that consistently and interpretably transform generated im…
Latent Compass: Creation by Navigation
Sarah Schwettmann, Hendrik Strobelt, Mauro Martino
In Marius von Senden's Space and Sight, a newly sighted blind patient describes the experience of a corner as lemon-like, because corners "prick" sight like lemons prick the tongue…