Picture for Steve Yadlowsky

Steve Yadlowsky

Gemini: A Family of Highly Capable Multimodal Models

Add code
Dec 19, 2023
Viaarxiv icon

Pretraining Data Mixtures Enable Narrow Model Selection Capabilities in Transformer Models

Nov 01, 2023
Figure 1 for Pretraining Data Mixtures Enable Narrow Model Selection Capabilities in Transformer Models
Figure 2 for Pretraining Data Mixtures Enable Narrow Model Selection Capabilities in Transformer Models
Figure 3 for Pretraining Data Mixtures Enable Narrow Model Selection Capabilities in Transformer Models
Figure 4 for Pretraining Data Mixtures Enable Narrow Model Selection Capabilities in Transformer Models
Viaarxiv icon

Choosing a Proxy Metric from Past Experiments

Sep 14, 2023
Figure 1 for Choosing a Proxy Metric from Past Experiments
Figure 2 for Choosing a Proxy Metric from Past Experiments
Figure 3 for Choosing a Proxy Metric from Past Experiments
Figure 4 for Choosing a Proxy Metric from Past Experiments
Viaarxiv icon

Diagnosing Model Performance Under Distribution Shift

Add code
Mar 10, 2023
Figure 1 for Diagnosing Model Performance Under Distribution Shift
Figure 2 for Diagnosing Model Performance Under Distribution Shift
Figure 3 for Diagnosing Model Performance Under Distribution Shift
Figure 4 for Diagnosing Model Performance Under Distribution Shift
Viaarxiv icon

Boosting the interpretability of clinical risk scores with intervention predictions

Add code
Jul 06, 2022
Figure 1 for Boosting the interpretability of clinical risk scores with intervention predictions
Figure 2 for Boosting the interpretability of clinical risk scores with intervention predictions
Figure 3 for Boosting the interpretability of clinical risk scores with intervention predictions
Figure 4 for Boosting the interpretability of clinical risk scores with intervention predictions
Viaarxiv icon

Calibration Error for Heterogeneous Treatment Effects

Add code
Mar 24, 2022
Figure 1 for Calibration Error for Heterogeneous Treatment Effects
Figure 2 for Calibration Error for Heterogeneous Treatment Effects
Figure 3 for Calibration Error for Heterogeneous Treatment Effects
Figure 4 for Calibration Error for Heterogeneous Treatment Effects
Viaarxiv icon

Evaluating Treatment Prioritization Rules via Rank-Weighted Average Treatment Effects

Nov 15, 2021
Figure 1 for Evaluating Treatment Prioritization Rules via Rank-Weighted Average Treatment Effects
Figure 2 for Evaluating Treatment Prioritization Rules via Rank-Weighted Average Treatment Effects
Figure 3 for Evaluating Treatment Prioritization Rules via Rank-Weighted Average Treatment Effects
Figure 4 for Evaluating Treatment Prioritization Rules via Rank-Weighted Average Treatment Effects
Viaarxiv icon

The MultiBERTs: BERT Reproductions for Robustness Analysis

Add code
Jun 30, 2021
Figure 1 for The MultiBERTs: BERT Reproductions for Robustness Analysis
Figure 2 for The MultiBERTs: BERT Reproductions for Robustness Analysis
Figure 3 for The MultiBERTs: BERT Reproductions for Robustness Analysis
Figure 4 for The MultiBERTs: BERT Reproductions for Robustness Analysis
Viaarxiv icon

Counterfactual Invariance to Spurious Correlations: Why and How to Pass Stress Tests

Jun 02, 2021
Figure 1 for Counterfactual Invariance to Spurious Correlations: Why and How to Pass Stress Tests
Figure 2 for Counterfactual Invariance to Spurious Correlations: Why and How to Pass Stress Tests
Figure 3 for Counterfactual Invariance to Spurious Correlations: Why and How to Pass Stress Tests
Figure 4 for Counterfactual Invariance to Spurious Correlations: Why and How to Pass Stress Tests
Viaarxiv icon

SLOE: A Faster Method for Statistical Inference in High-Dimensional Logistic Regression

Add code
Mar 23, 2021
Figure 1 for SLOE: A Faster Method for Statistical Inference in High-Dimensional Logistic Regression
Figure 2 for SLOE: A Faster Method for Statistical Inference in High-Dimensional Logistic Regression
Figure 3 for SLOE: A Faster Method for Statistical Inference in High-Dimensional Logistic Regression
Figure 4 for SLOE: A Faster Method for Statistical Inference in High-Dimensional Logistic Regression
Viaarxiv icon