95 citations · 238 across the 17 of their papers we have counts for
18 papers
D-Scope: Decomposing and Steering Diffusion Transformers with Sparse Autoencoders
Xinyue Xu, Jiahao Zhang, Lijie Hu +2
Sparse autoencoders (SAEs) reveal visual structure in diffusion transformers (DiTs), but interpreting a feature does not establish whether it can be used to control generation. We…
Can Sensitive Information Be Deleted From LLMs? Objectives for Defending Against Extraction Attacks
Vaidehi Patil, Peter Hase, Mohit Bansal
Pretrained language models sometimes possess knowledge that we do not wish them to, including memorized personal information and knowledge that could be used to harm people. They c…
Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback
Stephen Casper, Xander Davies, Claudia Shi +29
Reinforcement learning from human feedback (RLHF) is a technique for training AI systems to align with human goals. RLHF has emerged as the central method used to finetune state-of…
Adaptive Contextual Perception: How to Generalize to New Backgrounds and Ambiguous Objects
Zhuofan Ying, Peter Hase, Mohit Bansal
Biological vision systems make adaptive use of context to recognize objects in new settings with novel contexts as well as occluded or blurry objects in familiar settings. In this…
Does Localization Inform Editing? Surprising Differences in Causality-Based Localization vs. Knowledge Editing in Language Models
Peter Hase, Mohit Bansal, Been Kim +1
Language models learn a great quantity of factual information during pretraining, and recent work localizes this information to specific model weights like mid-layer MLP weights. I…
Are Hard Examples also Harder to Explain? A Study with Human and Model-Generated Explanations
Swarnadeep Saha, Peter Hase, Nazneen Rajani +1
Recent work on explainable NLP has shown that few-shot prompting can enable large pretrained language models (LLMs) to generate grammatical and factual natural language explanation…
Summarization Programs: Interpretable Abstractive Summarization with Neural Modular Trees
Swarnadeep Saha, Shiyue Zhang, Peter Hase +1
Current abstractive summarization models either suffer from a lack of clear interpretability or provide incomplete rationales by only highlighting parts of the source document. To…
VisFIS: Visual Feature Importance Supervision with Right-for-the-Right-Reason Objectives
Zhuofan Ying, Peter Hase, Mohit Bansal
Many past works aim to improve visual reasoning in models by supervising feature importance (estimated by model explanation techniques) with human annotations such as highlights of…
GrIPS: Gradient-free, Edit-based Instruction Search for Prompting Large Language Models
Archiki Prasad, Peter Hase, Xiang Zhou +1
Providing natural language instructions in prompts is a useful new paradigm for improving task performance of large language models in a zero-shot setting. Recent work has aimed to…
Do Language Models Have Beliefs? Methods for Detecting, Updating, and Visualizing Model Beliefs
Peter Hase, Mona Diab, Asli Celikyilmaz +5
Do language models have beliefs about the world? Dennett (1995) famously argues that even thermostats have beliefs, on the view that a belief is simply an informational state decou…
Low-Cost Algorithmic Recourse for Users With Uncertain Cost Functions
Prateek Yadav, Peter Hase, Mohit Bansal
People affected by machine learning model decisions may benefit greatly from access to recourses, i.e. suggestions about what features they could change to receive a more favorable…
The Out-of-Distribution Problem in Explainability and Search Methods for Feature Importance Explanations
Peter Hase, Harry Xie, Mohit Bansal
Feature importance (FI) estimates are a popular form of explanation, and they are commonly created and evaluated by computing the change in model confidence caused by removing cert…
When Can Models Learn From Explanations? A Formal Framework for Understanding the Roles of Explanation Data
Peter Hase, Mohit Bansal
Many methods now exist for conditioning model outputs on task instructions, retrieved documents, and user-provided explanations and feedback. Rather than relying solely on examples…
FastIF: Scalable Influence Functions for Efficient Model Interpretation and Debugging
Han Guo, Nazneen Fatema Rajani, Peter Hase +2
Influence functions approximate the "influences" of training data-points for test predictions and have a wide variety of applications. Despite the popularity, their computational c…
Leakage-Adjusted Simulatability: Can Models Generate Non-Trivial Explanations of Their Behavior in Natural Language?
Peter Hase, Shiyue Zhang, Harry Xie +1
Data collection for natural language (NL) understanding tasks has increasingly included human explanations alongside data points, allowing past works to introduce models that both…
Evaluating Explainable AI: Which Algorithmic Explanations Help Users Predict Model Behavior?
Peter Hase, Mohit Bansal
Algorithmic approaches to interpreting machine learning models have proliferated in recent years. We carry out human subject tests that are the first of their kind to isolate the e…
Interpretable Image Recognition with Hierarchical Prototypes
Peter Hase, Chaofan Chen, Oscar Li +1
Vision models are interpretable when they classify objects on the basis of features that a person can directly understand. Recently, methods relying on visual feature prototypes ha…
Shall I Compare Thee to a Machine-Written Sonnet? An Approach to Algorithmic Sonnet Generation
John Benhardt, Peter Hase, Liuyi Zhu +1
We provide an approach for generating beautiful poetry. Our sonnet-generation algorithm includes several novel elements that improve over the state of the art, leading to metrical,…