Picture for Yann Dubois

Yann Dubois

Learning to (Learn at Test Time): RNNs with Expressive Hidden States

Add code
Jul 05, 2024
Viaarxiv icon

Length-Controlled AlpacaEval: A Simple Way to Debias Automatic Evaluators

Add code
Apr 06, 2024
Viaarxiv icon

Identifying the Risks of LM Agents with an LM-Emulated Sandbox

Add code
Sep 25, 2023
Figure 1 for Identifying the Risks of LM Agents with an LM-Emulated Sandbox
Figure 2 for Identifying the Risks of LM Agents with an LM-Emulated Sandbox
Figure 3 for Identifying the Risks of LM Agents with an LM-Emulated Sandbox
Figure 4 for Identifying the Risks of LM Agents with an LM-Emulated Sandbox
Viaarxiv icon

AlpacaFarm: A Simulation Framework for Methods that Learn from Human Feedback

Add code
May 22, 2023
Figure 1 for AlpacaFarm: A Simulation Framework for Methods that Learn from Human Feedback
Figure 2 for AlpacaFarm: A Simulation Framework for Methods that Learn from Human Feedback
Figure 3 for AlpacaFarm: A Simulation Framework for Methods that Learn from Human Feedback
Figure 4 for AlpacaFarm: A Simulation Framework for Methods that Learn from Human Feedback
Viaarxiv icon

Evaluating Self-Supervised Learning via Risk Decomposition

Add code
Feb 06, 2023
Figure 1 for Evaluating Self-Supervised Learning via Risk Decomposition
Figure 2 for Evaluating Self-Supervised Learning via Risk Decomposition
Figure 3 for Evaluating Self-Supervised Learning via Risk Decomposition
Figure 4 for Evaluating Self-Supervised Learning via Risk Decomposition
Viaarxiv icon

Improving Self-Supervised Learning by Characterizing Idealized Representations

Add code
Sep 13, 2022
Figure 1 for Improving Self-Supervised Learning by Characterizing Idealized Representations
Figure 2 for Improving Self-Supervised Learning by Characterizing Idealized Representations
Figure 3 for Improving Self-Supervised Learning by Characterizing Idealized Representations
Figure 4 for Improving Self-Supervised Learning by Characterizing Idealized Representations
Viaarxiv icon

Is a Caption Worth a Thousand Images? A Controlled Study for Representation Learning

Add code
Jul 15, 2022
Figure 1 for Is a Caption Worth a Thousand Images? A Controlled Study for Representation Learning
Figure 2 for Is a Caption Worth a Thousand Images? A Controlled Study for Representation Learning
Figure 3 for Is a Caption Worth a Thousand Images? A Controlled Study for Representation Learning
Figure 4 for Is a Caption Worth a Thousand Images? A Controlled Study for Representation Learning
Viaarxiv icon

Learning Instance-Specific Data Augmentations

Add code
May 31, 2022
Figure 1 for Learning Instance-Specific Data Augmentations
Figure 2 for Learning Instance-Specific Data Augmentations
Figure 3 for Learning Instance-Specific Data Augmentations
Figure 4 for Learning Instance-Specific Data Augmentations
Viaarxiv icon

Optimal Representations for Covariate Shift

Add code
Dec 31, 2021
Figure 1 for Optimal Representations for Covariate Shift
Figure 2 for Optimal Representations for Covariate Shift
Figure 3 for Optimal Representations for Covariate Shift
Figure 4 for Optimal Representations for Covariate Shift
Viaarxiv icon

Lossy Compression for Lossless Prediction

Add code
Jul 07, 2021
Figure 1 for Lossy Compression for Lossless Prediction
Figure 2 for Lossy Compression for Lossless Prediction
Figure 3 for Lossy Compression for Lossless Prediction
Figure 4 for Lossy Compression for Lossless Prediction
Viaarxiv icon