Picture for Tom Joy

Tom Joy

What Makes and Breaks Safety Fine-tuning? A Mechanistic Study

Add code
Jul 16, 2024
Viaarxiv icon

What Makes and Breaks Safety Fine-tuning? Mechanistic Study

Add code
Jul 14, 2024
Viaarxiv icon

MoCaE: Mixture of Calibrated Experts Significantly Improves Object Detection

Add code
Sep 27, 2023
Viaarxiv icon

Towards Building Self-Aware Object Detectors via Reliable Uncertainty Quantification and Calibration

Add code
Jul 03, 2023
Viaarxiv icon

Sample-dependent Adaptive Temperature Scaling for Improved Calibration

Add code
Jul 22, 2022
Figure 1 for Sample-dependent Adaptive Temperature Scaling for Improved Calibration
Figure 2 for Sample-dependent Adaptive Temperature Scaling for Improved Calibration
Figure 3 for Sample-dependent Adaptive Temperature Scaling for Improved Calibration
Figure 4 for Sample-dependent Adaptive Temperature Scaling for Improved Calibration
Viaarxiv icon

Learning Multimodal VAEs through Mutual Supervision

Add code
Jul 01, 2021
Figure 1 for Learning Multimodal VAEs through Mutual Supervision
Figure 2 for Learning Multimodal VAEs through Mutual Supervision
Figure 3 for Learning Multimodal VAEs through Mutual Supervision
Figure 4 for Learning Multimodal VAEs through Mutual Supervision
Viaarxiv icon

Rethinking Semi-Supervised Learning in VAEs

Add code
Jun 17, 2020
Figure 1 for Rethinking Semi-Supervised Learning in VAEs
Figure 2 for Rethinking Semi-Supervised Learning in VAEs
Figure 3 for Rethinking Semi-Supervised Learning in VAEs
Figure 4 for Rethinking Semi-Supervised Learning in VAEs
Viaarxiv icon