Showing cs.LGShow all
3 papers · 1 filter
cs.LG2025
Subliminal Corruption: Mechanisms, Thresholds, and Interpretability
Reya Vir, Sarvesh Bhatnagar
As machine learning models are increasingly fine-tuned on synthetic data, there is a critical risk of subtle misalignments spreading through interconnected AI systems. This paper i…
cs.LG2024
Polynomial Regression as a Task for Understanding In-context Learning Through Finetuning and Alignment
Max Wilcoxson, Morten Svendgård, Ria Doshi +3
Simple function classes have emerged as toy problems to better understand in-context-learning in transformer-based architectures used for large language models. But previously prop…
cs.LG2023
Can Transformers Learn Sequential Function Classes In Context?
Ryan Campbell, Emma Guo, Evan Hu +2
In-context learning (ICL) has revolutionized the capabilities of transformer models in NLP. In our project, we extend the understanding of the mechanisms underpinning ICL by explor…