research
∙
07/18/2023
Overthinking the Truth: Understanding how Language Models Process False Demonstrations
Modern language models can imitate complex patterns through few-shot lea...
share
research
∙
07/05/2023
Jailbroken: How Does LLM Safety Training Fail?
Large language models trained for safety and harmlessness remain suscept...
research
∙
06/26/2023
Improved Bayes Risk Can Yield Reduced Social Welfare Under Competition
As the scale of machine learning models increases, trends such as scalin...
research
∙
06/21/2023
Mass-Producing Failures of Multimodal Systems with Language Models
Deployed multimodal systems can fail in ways that evaluators did not ant...
research
∙
06/13/2023
Incentivizing High-Quality Content in Online Recommender Systems
For content recommender systems such as TikTok and YouTube, the platform...
research
∙
03/14/2023
Eliciting Latent Predictions from Transformers with the Tuned Lens
We analyze transformers from the perspective of iterative inference, see...
research
∙
03/08/2023
Automatically Auditing Large Language Models via Discrete Optimization
Auditing large language models for unexpected behaviors is critical to p...
research
∙
02/28/2023
Goal Driven Discovery of Distributional Differences via Language Descriptions
Mining large corpora can generate useful discoveries but is time-consumi...
research
∙
02/23/2023
Reward Learning as Doubly Nonparametric Bandits: Optimal Design and Scaling Laws
Specifying reward functions for complex tasks like object manipulation o...
research
∙
01/12/2023
Progress measures for grokking via mechanistic interpretability
Neural networks often exhibit emergent behavior, where qualitatively new...
research
∙
12/07/2022
Discovering Latent Knowledge in Language Models Without Supervision
Existing techniques for training language models can be misaligned with ...
research
∙
11/01/2022
Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small
Research in mechanistic interpretability seeks to explain behaviors of m...
research
∙
10/18/2022
How Would The Viewer Feel? Estimating Wellbeing From Video Scenarios
In recent years, deep neural networks have demonstrated increasingly str...
research
∙
06/30/2022
Forecasting Future World Events with Neural Networks
Forecasting future world events is a challenging but valuable task. Fore...
research
∙
06/27/2022
Auditing Visualizations: Transparency Methods Struggle to Detect Anomalous Behavior
Transparency methods such as model visualizations provide information th...
research
∙
06/27/2022
Supply-Side Equilibria in Recommender Systems
Digital recommender systems such as Spotify and Netflix affect not only ...
research
∙
02/24/2022
Capturing Failures of Large Language Models via Human Cognitive Biases
Large language models generate complex, open-ended outputs: instead of o...
research
∙
02/11/2022
Predicting Out-of-Distribution Error with the Projection Norm
We propose a metric – Projection Norm – to predict a model's performance...
research
∙
01/28/2022
Summarizing Differences between Text Distributions with Natural Language
How do two distributions of texts differ? Humans are slow at answering t...
research
∙
01/10/2022
The Effects of Reward Misspecification: Mapping and Mitigating Misaligned Models
Reward hacking – where RL agents exploit gaps in misspecified reward fun...
research
∙
12/09/2021
PixMix: Dreamlike Pictures Comprehensively Improve Safety Measures
In real-world applications of machine learning, reliable and safe system...
research
∙
12/08/2021
The Effect of Model Size on Worst-Group Generalization
Overparameterization is shown to result in poor test accuracy on rare su...
research
∙
10/25/2021
What Would Jiminy Cricket Do? Towards Agents That Behave Morally
When making everyday decisions, people are guided by their conscience, a...
research
∙
09/28/2021
Unsolved Problems in ML Safety
Machine learning (ML) systems are rapidly increasing in size, are acquir...
research
∙
08/19/2021
Learning Equilibria in Matching Markets from Bandit Feedback
Large-scale, two-sided matching platforms must find market outcomes that...
research
∙
08/03/2021
Grounding Representation Similarity with Statistical Testing
To understand neural network behavior, recent works quantitatively compa...
research
∙
05/20/2021
Measuring Coding Challenge Competence With APPS
While programming is one of the most broadly applicable skills in modern...
research
∙
05/13/2021
Are Larger Pretrained Language Models Uniformly Better? Comparing Performance at the Instance Level
Larger language models have higher accuracy on average, but are they bet...
research
∙
04/17/2021
Agnostic learning with unknown utilities
Traditional learning approaches for classification implicitly assume tha...
research
∙
03/17/2021
Understanding Generalization in Adversarial Training via the Bias-Variance Decomposition
Adversarially trained models exhibit a large generalization gap: they ca...
research
∙
03/13/2021
Approximating How Single Head Attention Learns
Why do models often attend to salient words, and how does this evolve th...
research
∙
03/10/2021
Limitations of Post-Hoc Feature Alignment for Robustness
Feature alignment is an approach to improving robustness to distribution...
research
∙
03/05/2021
Measuring Mathematical Problem Solving With the MATH Dataset
Many intellectual endeavors require mathematical problem solving, but th...
research
∙
10/22/2020
Enabling certification of verification-agnostic networks via memory-efficient semidefinite programming
Convex relaxations have emerged as a promising approach for verifying de...
research
∙
09/07/2020
Measuring Massive Multitask Language Understanding
We propose a new test to measure a text model's multitask accuracy. The ...
research
∙
08/05/2020
Aligning AI With Shared Human Values
We show how to assess a language model's knowledge of basic concepts of ...
research
∙
06/29/2020
The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization
We introduce three new robustness benchmarks consisting of naturally occ...
research
∙
05/28/2020
Robust estimation via generalized quasi-gradients
We explore why many recently proposed robust estimation problems are eff...
research
∙
05/19/2020
Identifying Statistical Bias in Dataset Replication
Dataset replication is a useful tool for assessing whether improvements ...
research
∙
02/26/2020
Rethinking Bias-Variance Trade-off for Generalization of Neural Networks
The classical bias-variance trade-off predicts that bias decreases and v...
research
∙
01/21/2020
When does the Tukey median work?
We analyze the performance of the Tukey median estimator under total var...
research
∙
11/25/2019
A Benchmark for Anomaly Segmentation
Detecting out-of-distribution examples is important for safety-critical ...
research
∙
09/19/2019
Generalized Resilience and Robust Statistics
Robust statistics traditionally focuses on outliers, or perturbations in...
research
∙
08/21/2019
Testing Robustness Against Unforeseen Adversaries
Considerable work on adversarial defense has studied robustness to a fix...
research
∙
07/16/2019
Natural Adversarial Examples
We introduce natural adversarial examples -- real-world, unmodified, and...
research
∙
05/03/2019
Transfer of Adversarial Robustness Between Perturbation Types
We study the transfer of adversarial robustness of deep neural networks ...
research
∙
11/13/2018
FrAngel: Component-Based Synthesis with Control Structures
In component-based program synthesis, the synthesizer generates a progra...
research
∙
11/02/2018
Semidefinite relaxations for certifying robustness to adversarial examples
Despite their impressive performance on diverse tasks, neural networks f...
research
∙
11/02/2018
Stronger Data Poisoning Attacks Break Data Sanitization Defenses
Machine learning models trained on data from the outside world can be co...
research
∙
07/09/2018