Ehsan Abbasnejad

23

Papers

43

Total Citations

Papers (23)

Neural Redshift: Random Networks are not Random Functions

Towards Higher Effective Rank in Parameter-Efficient Fine-tuning using Khatri-Rao Product

Bayesian Low-Rank Learning (Bella): A Practical Approach to Bayesian Neural Networks

Synergy and Diversity in CLIP: Enhancing Performance Through Adaptive Backbone Ensembling

Gold Seeker: Information Gain From Policy Distributions for Goal-Oriented Vision-and-Langauge Reasoning

Counterfactual Vision and Language Learning

All Labels Are Not Created Equal: Enhancing Semi-Supervision via Label Grouping and Co-Training

Evading the Simplicity Bias: Training a Diverse Set of Models Discovers Solutions With Superior OOD Generalization

Active Learning by Feature Mixing

ProtoCon: Pseudo-Label Refinement via Online Clustering and Prototypical Consistency for Efficient Semi-Supervised Learning

DeepSetNet: Predicting Sets With Deep Neural Networks

Unshuffling Data for Improved Generalization in Visual Question Answering

Energy-based Self-Training and Normalization for Unsupervised Domain Adaptation

Predicting Is Not Understanding: Recognizing and Addressing Underspecification in Machine Learning

Do We Always Need the Simplicity Bias? Looking for Optimal Inductive Biases in the Wild

Mysteries of the Deep: Role of Intermediate Representations in Out of Distribution Detection

Selective Mixup Helps with Distribution Shifts, But Not (Only) because of Mixup

What's to Know? Uncertainty as a Guide to Asking Goal-Oriented Questions

On the Value of Out-of-Distribution Testing: An Example of Goodhart's Law

Counterfactual Vision-and-Language Navigation: Unravelling the Unseen

Truncated Matrix Power Iteration for Differentiable DAG Learning

RanPAC: Random Projections and Pre-trained Models for Continual Learning

ID and OOD Performance Are Sometimes Inversely Correlated on Real-world Datasets