5 citations · 13 across the 15 of their papers we have counts for
18 papers
A Phenomenology of Neural Geometry in Vision-Language-Action Models
Mozes Jacobs, Aiden Swann, Mathilde Papillon +16
Robotic manipulation is inherently geometric, but what shape does the world take inside the policy that performs it? Prior work on vision-language-action (VLA) policies establishes…
Multi-Dimensional Featurizers Yield Fine-Grained Steering and Data Shaping in Robot Foundation Models
Aiden Swann, Mozes Jacobs, Mathilde Papillon +12
Can better feature discovery improve control over robot behavior through internal interventions and training data? Recent work in vision and language models reveals multidimensiona…
Language Model Activations Inhabit Privileged Error-Correcting Basins
Matthew Finlayson, Francisco Pernice, Eric Todd +16
Language models exhibit remarkable robustness, continuing to produce coherent text even when their activations are perturbed by interventions like linear steering. We hypothesize t…
Structuring Sparsity: Block-Sparse Featurizers Capture Visual Concept Manifolds
Thomas Fel, Matthew Kowal, Mozes Jacobs +22
What is the geometry of a visual percept? The most widely used protocols for decomposing neural network representations into interpretable parts treat concepts as isolated directio…
Anatomy of Post-Training: Using Interpretability to Characterize Data and Shape the Learning Signal
Leon Bergen, Usha Bhalla, Sidharth Baskaran +14
Language-model post-training is the main stage at which model behavior is shaped, yet it still largely involves optimization of scalar rewards that summarize diverse desiderata. Th…
Manifold Steering Reveals the Shared Geometry of Neural Network Representation and Behavior
Daniel Wurgaft, Can Rager, Matthew Kowal +13
Neural representations carry rich geometric structure; but does that structure causally shape behavior? To address this question, we intervene along paths through activation space…
Arithmetic in the Wild: Llama uses Base-10 Addition to Reason About Cyclic Concepts
Sheridan Feucht, Tal Haklay, Usha Bhalla +9
Does structure in representations imply structure in computation? We study how Llama-3.1-8B reasons over cyclic concepts (e.g., "what month is six months after August?"). Even thou…
Do Sparse Autoencoders Capture Concept Manifolds?
Usha Bhalla, Thomas Fel, Can Rager +9
Sparse autoencoders (SAEs) are widely used to extract interpretable features from neural network representations, often under the implicit assumption that concepts correspond to in…
RippleBench: Capturing Ripple Effects Using Existing Knowledge Repositories
Roy Rinberg, Usha Bhalla, Igor Shilov +2
Targeted interventions on language models, such as unlearning or model editing, aim to modify specific information, but their effects often propagate to related, unintended areas (…
Temporal Sparse Autoencoders: Leveraging the Sequential Nature of Language for Interpretability
Usha Bhalla, Alex Oesterling, Claudio Mayrink Verdun +2
Translating the internal representations and computations of models into concepts that humans can understand is a key goal of interpretability. While recent dictionary learning met…
Evaluating Adversarial Robustness of Concept Representations in Sparse Autoencoders
Aaron J. Li, Suraj Srinivas, Usha Bhalla +1
Sparse autoencoders (SAEs) are commonly used to interpret the internal activations of large language models (LLMs) by mapping them to human-interpretable concept representations. W…
Towards Unified Attribution in Explainable AI, Data-Centric AI, and Mechanistic Interpretability
Shichang Zhang, Tessa Han, Usha Bhalla +1
The increasing complexity of AI systems has made understanding their behavior critical. Numerous interpretability methods have been developed to attribute model behavior to three k…
Towards Unifying Interpretability and Control: Evaluation via Intervention
Usha Bhalla, Suraj Srinivas, Asma Ghandeharioun +1
With the growing complexity and capability of large language models, a need to understand model reasoning has emerged, often motivated by an underlying goal of controlling and alig…
All Roads Lead to Rome? Exploring Representational Similarities Between Latent Spaces of Generative Image Models
Charumathi Badrinath, Usha Bhalla, Alex Oesterling +2
Do different generative image models secretly learn similar underlying representations? We investigate this by measuring the latent space similarity of four different models: VAEs,…
Operationalizing the Blueprint for an AI Bill of Rights: Recommendations for Practitioners, Researchers, and Policy Makers
Alex Oesterling, Usha Bhalla, Suresh Venkatasubramanian +1
As Artificial Intelligence (AI) tools are increasingly employed in diverse real-world applications, there has been significant interest in regulating these tools. To this end, seve…
Interpreting CLIP with Sparse Linear Concept Embeddings (SpLiCE)
Usha Bhalla, Alex Oesterling, Suraj Srinivas +2
CLIP embeddings have demonstrated remarkable performance across a wide range of multimodal applications. However, these high-dimensional, dense vector representations are not easil…
Discriminative Feature Attributions: Bridging Post Hoc Explainability and Inherent Interpretability
Usha Bhalla, Suraj Srinivas, Himabindu Lakkaraju
With the increased deployment of machine learning models in various real-world applications, researchers and practitioners alike have emphasized the need for explanations of model…
Do Vision-Language Pretrained Models Learn Composable Primitive Concepts?
Tian Yun, Usha Bhalla, Ellie Pavlick +1
Vision-language (VL) pretrained models have achieved impressive performance on multimodal reasoning and zero-shot recognition tasks. Many of these VL models are pretrained on unlab…