6 papers · 1 filter
How Do LLMs Use Their Depth?
Akshat Gupta, Jay Yeung, Gopala Anumanchipalli +1
Growing evidence suggests that large language models do not use their depth uniformly, yet we still lack a fine-grained understanding of their layer-wise prediction dynamics. In th…
The Oracle Has Spoken: A Multi-Aspect Evaluation of Dialogue in Pythia
Zixun Chen, Petr Babkin, Akshat Gupta +2
Dialogue is one of the landmark abilities of large language models (LLMs). Despite its ubiquity, few studies actually distinguish specific ingredients underpinning dialogue behavio…
Disentangling Codemixing in Chats: The NUS ABC Codemixed Corpus
Svetlana Churina, Akshat Gupta, Insyirah Mujtahid +1
Code-mixing involves the seamless integration of linguistic elements from multiple languages within a single discourse, reflecting natural multilingual communication patterns. Desp…
Efficient Knowledge Editing via Minimal Precomputation
Akshat Gupta, Maochuan Lu, Thomas Hartvigsen +1
Knowledge editing methods like MEMIT are able to make data and compute efficient updates of factual knowledge by using a single sentence to update facts and their consequences. How…
Norm Growth and Stability Challenges in Localized Sequential Knowledge Editing
Akshat Gupta, Christine Fang, Atahan Ozdemir +4
This study investigates the impact of localized updates to large language models (LLMs), specifically in the context of knowledge editing - a task aimed at incorporating or modifyi…
Lifelong Knowledge Editing requires Better Regularization
Akshat Gupta, Phudish Prateepamornkul, Maochuan Lu +3
Knowledge editing is a promising way to improve factuality in large language models, but recent studies have shown significant model degradation during sequential editing. In this…